Special thanks to Tarun Chitra(Gauntlet), Nate Hindman(Bio), and Benji Leibowitz(Molecule) for feedback and review

I recently earned a Ph.D. in Chemical Engineering and published four first-author papers during my studies. Among these were publications in some of the highest-ranking academic journals, including Nature's sister journals and the Journal of the American Chemical Society (JACS).

Although my academic experience has been limited to that of a graduate student without serving as a principal investigator, which could be an incomplete perspective, my nearly six years in academia noticed numerous structural issues within the system.

In this context, the idea of DeSci (Decentralized Science) leveraging blockchain technology to challenge centralized structures in science is undoubtedly fascinating. The crypto market has recently been swept by a DeSci trend, with many claiming it could revolutionize the scientific landscape.

I, too, hope for such a transformation. However, I believe that the chance of DeSci completely overturning traditional academia is not high. To summarize my view, the most likely scenario is that DeSci will play a complementary role in addressing specific issues within the conventional academic system.

Thus, with all the recent enthusiasm for DeSci, I would like to take this opportunity to explore some of the structural issues in traditional academia based on my brief experience, evaluate whether blockchain technology can genuinely address these issues, and discuss the potential impact of DeSci on the academic world.

1. The Sudden DeSci Fever

1.1 DeSci: From a Niche Concept to a Growing Movement

The longstanding structural issues within academia have been well-documented, as seen in articles like VOX’s “The 7 biggest problems facing science, according to 270 scientists” and “The war to free science.” Over the years, there have been numerous attempts to address these challenges, some of which will be explored later.

The concept of DeSci, which seeks to solve these problems by incorporating blockchain technology into scientific research, only began to gain attention around 2020. Brian Armstrong, the CEO of Coinbase, introduced the idea to the crypto community through ResearchHub, aiming to realign incentives in science via ResearchCoin (RSC).

However, due to the speculative nature of capital in the crypto market, DeSci failed to attract widespread interest among users. For a long time, only small communities championed its future—until the emergence of pump.science.

1.2 The Butterfly Effect of pump.science

(Source: pump.science)

pump.science is a DeSci project in the Solana ecosystem built by Molecule, a well-known DeSci platform. It functions as a funding platform while streaming long-term experiments using Wormbot technology. Users can propose compounds they believe could extend their lifespan or purchase tokens associated with these ideas.

Once the token’s market cap surpasses a certain threshold, experiments are conducted using Wormbot equipment to verify whether the compound can truly extend the lifespan of the test subjects. If successful, token holders gain rights to the compound. (However, some community members have criticized this approach, claiming that the experiments lack sufficient scientific rigor and are unlikely to lead to actual life-extending pharmaceuticals. Gwart's sarcastic remark reflects a particular school of thought that eyes DeSci with skepticism and questions the arguments made by proponents.)

pump.science adopted the bonding curve mechanism, similar to what Molecule uses, meaning the token price increases as more users purchase it. The launch of tokens like RIF (representing Rifampicin) and URO (representing Urolithin A) coincided with a meme token frenzy in the crypto market, driving their prices higher. This price surge unintentionally brought widespread attention to DeSci. Ironically, it was not DeSci’s essence but the speculative rise in token prices that ignited the current wave of interest in DeSci.

(Source: Kaito)

In the fast-moving crypto market, where DeSci had long been a niche sector, November 2024 saw it become one of the hottest narratives. Not only did the tokens from pump.science skyrocket, but Binance announced its investment in DeSci funding protocol Bio, while other established DeSci tokens also experienced significant price increases, marking a pivotal moment for the movement.

2. Where Traditional Science Falls Short

No exaggeration—academia faces numerous systemic and severe issues. During my time in the academic world, I constantly questioned how such a flawed structure could remain sustainable. Before diving into the potential of DeSci, let’s first examine the shortcomings of the traditional academic system.

2.1 Systemic Challenge 1: Funding

2.1.1 The Evolution of R&D Funding

Before the 19th century, scientists secured research funding and earned their livings in very different ways from today:

• Patronage: European monarchs and aristocrats provided financial support to researchers to enhance their prestige and contribute to scientific advancements. For instance, Galileo received patronage from the Medici family, enabling him to continue his telescope development and astronomical studies. Religious institutions also played a role in advancing science, with churches and clergy funding research in astronomy, mathematics, and medicine during the Middle Ages.

• Self-Funding: Many scientists sustained their research through personal income from other professions. They worked as university professors, teachers, authors, or engineers to fund their scientific endeavors.

In the late 19th and early 20th centuries, centralized funding systems from governments and corporations began to take root. During World War I and II, governments established various agencies and invested heavily in defense research to secure victory in the wars.

In the U.S., organizations like the National Advisory Committee for Aeronautics (NACA) and the National Research Council (NRC) were founded during World War I. Similarly, in Germany, the predecessor of today’s German Research Foundation (DFG), Notgemeinschaft der Deutschen Wissenschaft, was established in 1920. Around the same time, corporate research labs like Bell Labs and GE Research also emerged, marking a shift where corporations joined governments in actively funding R&D.

This government- and corporate-driven funding model became the norm and continues to dominate today. Governments and corporations allocate significant budgets to R&D, supporting researchers worldwide. For instance, in 2023, the U.S. federal government spent a staggering $190 billion on R&D, a 13% increase compared to 2022.

(Source: ResearchHub)

In the United States, the funding process involves the federal government allocating a portion of its budget to R&D. These funds are then distributed to various agencies. Prominent examples include the National Institutes of Health (NIH), the largest funder of biomedical research; the Department of Defense (DoD), which focuses on defense research; the National Science Foundation (NSF), funding science and engineering across disciplines; the Department of Energy (DOE), responsible for renewable energy and nuclear physics; and NASA, which supports space and aeronautics research.

2.1.2 Centralized Funding Distorts Science

Today, it is virtually impossible for university professors to conduct research independently without external funding. As a result, they are forced to rely on financial support from governments or corporations. Many of the issues affecting modern academia arise from this centralized funding model.

The first major issue is inefficiency in the funding process. Although the process details differ by country and organization, it is universally described as lengthy, opaque, and inefficient.

To secure funding, research labs must go through extensive paperwork and presentations, undergoing rigorous evaluations by government or corporate bodies. While prestigious and well-established labs can receive millions or even tens of millions of dollars from a single grant, requiring less frequent involvement in the funding process, this is not the norm.

For most labs, funding is typically tens of thousands of dollars, necessitating repeated applications, extensive documentation, and continuous reviews. Conversations with graduate student friends show that many researchers and students cannot dedicate their time fully to research. Instead, they are consumed by tasks related to funding applications and participating in corporate projects.

Additionally, many of these corporate projects are minimally relevant to the students’ graduation research, underscoring this system's inefficiencies.

(Source: NSF)

Spending significant time on funding applications may eventually pay off, but unfortunately, securing funding is not easy. According to the NSF, the funding rates for 2023 and 2024 were 29% and 26%, respectively, with the median annual grant size being a modest $150,000. Similarly, the NIH reports funding success rates that typically range between 15% and 30%. Since a single grant is often insufficient for many academic researchers, they are forced to apply multiple times to sustain their work.

The challenges don’t stop there. Networking plays a crucial role in securing funding. Professors often collaborate with their peers rather than applying independently to increase their chances of obtaining grants. It’s also not uncommon for professors to engage in informal lobbying with funding stakeholders to secure corporate funding. This reliance on networking and the lack of transparency in the funding selection process are significant barriers for early-career researchers attempting to enter the system.

Another major issue with centralized funding is the lack of incentives for long-term research. Grants lasting more than five years are extremely rare. According to NSF data, most grants are awarded for 1–5 years, and other government agencies follow a similar pattern. Corporate R&D projects also typically provide grants for 1–3 years, depending on the company and project.

Politics heavily influence government funding. For example, during the Trump administration, defense R&D funding increased significantly, while under Democratic leadership, funding tended to focus on environmental research. Because government priorities shift with political agendas, long-term funding projects are uncommon.

Corporate funding faces similar limitations. In 2022, the median tenure of S&P 500 CEOs was 4.8 years, with other executives serving for comparable durations. Given that companies must adapt quickly to changing industries and technologies—and these executives often make funding decisions—corporate-funded projects rarely extend over long periods.

As a result, centralized funding systems incentivize researchers to pursue projects that yield quick and tangible results. To secure continuous funding, researchers are pressured to produce results within five years, leading them to select research topics that fit this timeline. This perpetuates a cycle of short-term focus, so only a handful of groups or institutions undertake long-term projects requiring more than five years.

Centralized funding also drives researchers toward producing a higher quantity of lower-quality work due to the pressure to deliver quick results. Research can be divided into incremental advancements that build slightly upon existing knowledge and groundbreaking discoveries that create entirely new territory. Centralized funding systems naturally prioritize the first option over the second. Most studies published in journals outside the top tier offer incremental improvements rather than transformative insights.

While it’s true that modern science has become highly specialized, making groundbreaking discoveries more challenging, centralized funding systems make the problem worse by further discouraging innovative research. This systemic preference for incremental work acts as yet another obstacle to revolutionary advancements in science.

(Source: Nature)

Some researchers even manipulate data or make false claims. The current funding mechanisms, which demand results within tight timeframes, create incentives for such misconduct. As a graduate student, it wasn’t uncommon to hear news of students from other labs falsifying data. According to Nature, the proportion of retracted papers in conferences and journals has sharply increased over time.

2.1.3 Don’t Be Misled: Centralized Funding is Inevitable

To clarify, centralized funding itself isn’t inherently bad. While this funding model has led to these negative side effects, it is essential for modern science. Unlike in the past, today’s scientific research is highly complex and sophisticated. A single research project by a graduate student can cost anywhere from thousands to hundreds of thousands of dollars, and large-scale efforts like defense, aerospace, or fundamental physics require exponentially more resources.

Centralized funding is essential, but the accompanying problems must be addressed.

2.2 Systemic Challenge 2: Journals

2.2.1 Journal Business Overview

Companies like Tether, Circle (stablecoin issuers), Binance, and Coinbase (centralized exchanges) are seen as dominant players in the crypto industry. Similarly, in academia, the most powerful entities are academic journals. Key examples include Elsevier, Springer Nature, Wiley, the American Chemical Society, and IEEE.

For instance, Elsevier generated $3.67 billion in revenue and $2.55 billion in net income in 2022, achieving an extraordinary net profit margin of nearly 70%. In perspective, Nvidia’s net profit margin hovered around 55–57% in 2024. Meanwhile, Springer Nature recorded $1.44 billion in revenue in the first nine months of 2024 alone, highlighting the massive scale of the academic publishing business.

The typical revenue streams for academic journals include:

• Subscription Fees: Accessing papers published in a journal often requires a subscription or a one-time fee to access specific articles.

• Article Processing Charges (APCs): Many papers are behind paywalls. However, authors can opt to pay publication costs to make their articles open-access.

• Licensing and Reprints: In most cases, copyright is transferred to the publisher upon publication. Journals monetize these papers through educational or commercial licensing.

2.2.2 Journals: The Epicenter of Misaligned Incentives

At this point, you might wonder, “Why are journals the apex predators of academia? Isn’t their business structure similar to other industries?” The answer is no. Journals exemplify misaligned incentives in academia.

While traditional publishers or online platforms typically aim to make authors’ work accessible to a broad audience and share revenue with the creators, academic journals are structured entirely in favor of the publishers.

Journals play a crucial role in communicating researchers’ findings to readers, but their revenue models are primarily designed to benefit the publishers, leaving authors and readers with minimal advantages.

Readers wishing to access articles from specific journals must pay subscription fees or purchase individual articles. However, if researchers want to publish their work as open-access, they must pay processing charges to the journals, and they do not receive any share of the revenue generated. It doesn’t stop there—researchers not only forgo revenue sharing but, in most cases, the copyright of their work is transferred to the journal upon publication, allowing the journal to monetize the content. This system is highly exploitative and fundamentally unfair to researchers.

The business model of journals is exploitative in its revenue flow and brutal in terms of scale. For example, one of the most prominent fully open-access journals in the natural sciences, Nature Communications, charges authors an exorbitant $6,790 per article as an article processing fee. Researchers are required to pay this amount to have their papers published in Nature Communications.

(Source: ACS)

The subscription fees for academic journals are also staggering. While annual institutional subscription fees vary depending on the journal's field and type, the average annual subscription fee for journals under the American Chemical Society (ACS) is $4,908 per journal. If an institution subscribes to all ACS journals, the cost rises to an astronomical $170,000. For journals under Springer Nature, the average annual subscription fee is around $10,000 per journal, and subscribing to all of their journals costs about $630,000. Since most research institutions subscribe to numerous journals, the subscription expenses for readers can be exceptionally high.

The most troubling aspect of this system is that researchers are effectively forced to publish in journals to build their academic credentials, and much of the money flowing through the journal business comes from government or corporate research funding:

• Researchers must consistently build their track record to secure funding and advance their careers. Journal publications are the most critical and often the only way to achieve this.

• The research conducted to write these papers is primarily funded by government or corporate grants.

• The processing charges for publishing open-access articles are also paid using these grants.

• Subscription fees paid by institutions for accessing journal articles are likewise covered by these grants.

Since researchers primarily use external funding rather than personal funds, they may be more inclined to accept these expenses. Academic journals have exploited this system by charging authors and readers while retaining the copyright of published work, creating an egregiously exploitative revenue model.

2.2.3 Poorly-Designed Peer Review Process

The problems with journals extend beyond their revenue structure to the inefficiencies and lack of transparency in their publishing processes. Over my six years in academia, during which I published four papers, I encountered many issues, particularly the inefficient submission process and the opaque, luck-dependent peer review system.

The standard peer review process for most journals typically follows these steps:

1. Researchers compile their findings into a manuscript and submit it to their target journal.

2. The journal editor evaluates whether the manuscript aligns with its scope and meets general standards. If deemed suitable, the editor assigns two to three peer reviewers to evaluate the paper.

3. Peer reviewers assess the manuscript, providing feedback through comments and questions. They then make one of four recommendations:

   • Accept: Approve the manuscript without revisions.

   • Minor Revisions: Approve the manuscript pending minor corrections.

   • Major Revisions: Approve the manuscript pending substantial changes.

   • Reject: Decline the manuscript outright.

4. The researcher revises the paper based on the reviewers’ feedback, after which the editor makes a final decision.

While seemingly straightforward, this process is fraught with inefficiencies, inconsistencies, and a significant reliance on subjective judgment, which can undermine the system's quality and fairness.

The first issue is the highly inefficient review process. While I cannot speak for other fields, in natural sciences and engineering, the timeline for submitting a paper and proceeding through the review process is roughly as follows:

• Time to receive an editor rejection after submission: 1 week to 2 months

• Time to receive peer reviews after submission: 3 weeks to 4 months

• Time to receive a final decision after submission: 3 months to 1 year

When delays occur due to the journal's or reviewers' circumstances and if multiple rounds of peer review are required, it can take more than a year to publish a paper. For instance, in my case, the editor sent my paper to three peer reviewers, but one did not respond. This required finding another reviewer extending the peer review process to four months.

Worse, if the paper is rejected after this lengthy process, the entire cycle must be repeated with another journal, doubling the time required. Such an inefficient and time-consuming publishing process can be detrimental to researchers, as similar studies by other groups may get published during this time. I have seen this happen often, and since novelty is one of the most critical aspects of a paper, this can lead to severe consequences for researchers.

The second issue is the shortage of peer reviewers. As mentioned earlier, a submitted paper is typically evaluated by two to three peer reviewers. Whether the paper is accepted or rejected largely depends on these few individuals' opinions. Although reviewers are experts in related fields, and consensus on the paper's quality is often reached, there is still an element of chance involved.

Let me illustrate with an example from my experience. I once submitted a paper to the prestigious Journal A. Despite receiving two major comments and one minor comment, my paper was rejected. I then submitted the same paper to Journal B, which is slightly less prestigious. However, it was rejected again after receiving one rejection and one major comment. Interestingly, the outcome was worse in Journal B despite it being less prominent than Journal A.

This highlights a problem: paper evaluations rely on a small number of experts, and the selection of reviewers is entirely at the discretion of the journal editor. This means there’s an element of luck in whether the paper is approved. In an extreme example, the same paper might be accepted if reviewed by three lenient reviewers but rejected if assigned to three critical ones.

That said, significantly increasing the number of peer reviewers for fairer evaluation isn’t practical. From the journal’s perspective, more reviewers mean more communication and inefficiencies.

The third issue is the lack of incentives in the peer review process, leading to low-quality comments. This varies depending on the peer reviewer. Some reviewers thoroughly understand the paper and provide thoughtful comments and questions. Others, however, do not read the paper carefully, ask about information already included, or give irrelevant criticism and comments, leading to major revisions or rejection. This is unfortunately common and can leave researchers feeling betrayed as if their efforts have been invalidated.

This stems from the absence of incentives for the peer review process, which makes quality control difficult. When journals receive submissions, editors typically ask university professors or researchers in related fields to review the papers. However, even if these individuals spend time reading, analyzing, and commenting on the papers, they are not rewarded for their efforts. From the perspective of professors or graduate students, peer reviewing is merely an unpaid, burdensome task.

The fourth issue is the lack of transparency in the peer review process. Peer reviews are conducted anonymously to ensure fairness, and the journal editor selects reviewers. However, reviewers can identify the authors of the papers they review. This can lead to biased evaluations, such as giving favorable reviews to papers from friendly researchers or deliberately harsh reviews for papers from competing groups. Such instances are more common than one might expect.

2.2.4 The Illusion of Impact Factor

The final issue I’d like to address regarding journals is citation counts. How can we evaluate the career and expertise of researchers? Each researcher has unique strengths: some excel at experimental design, others are skilled at identifying research topics, and some can thoroughly investigate overlooked details. However, it is practically impossible to assess every researcher qualitatively. As a result, academia relies on quantitative metrics, represented by a single number, to evaluate researchers—specifically, citation counts and the H-index.

Researchers with higher H-index scores and citation counts for their published papers are generally regarded as more accomplished. For context, the H-index is a metric that evaluates a researcher's productivity and impact. For example, an H-index of 10 means the researcher has at least 10 papers, each cited 10 times or more. Ultimately, citation counts remain the most important metric.

What can researchers do to increase their citation counts? While producing high-quality papers is the fundamental solution, selecting the right research topic is equally critical. The more popular the field of study and the larger the pool of researchers, the more likely it is that citation counts will increase naturally.

(Source: Clarivate)

The table above shows the 2024 Journal Impact Factor ranking published by Clarivate. The impact factor (IF) represents the average number of citations a paper in a particular journal receives yearly. For example, if a journal's impact factor is 10, a researcher publishing in that journal can expect their paper to receive approximately 10 citations per year.

Looking at the rankings, it becomes evident that journals with high impact factors are generally concentrated in certain fields of research. Examples include cancer, medicine, materials, energy, and machine learning. Even within a broader field like chemistry, specific subfields such as batteries and eco-friendly energy tend to have an advantage in citation counts compared to traditional areas like organic chemistry. This indicates a potential risk in academia, where researchers might gravitate toward specific topics due to the heavy reliance on citation counts as a primary evaluation method.

This highlights that metrics like citation counts and impact factors are not universal tools for assessing the quality of researchers or journals. For instance, within the same ACS publisher group, ACS Energy Letters has an impact factor of 19, while JACS has an impact factor of 14.4. However, JACS is considered one of the most prestigious and authoritative journals in the field of chemistry. Similarly, Nature is widely regarded as the top journal for researchers to publish in, yet its impact factor is 50.5 because it publishes papers on a wide range of topics. In contrast, Nature Medicine, a sister journal focusing on a specific field, has a higher impact factor of 58.7.

2.2.5 Publish or Perish

Success is born from failure. Progress in any domain requires failure as a stepping stone. The research findings published in academia today are often the result of countless hours and failed attempts. However, in modern scientific circles, almost all papers report only successful results, while the many failures leading up to those successes are left unpublished and discarded. In the competitive world of academia, researchers have little incentive to report failed experiments as they offer no benefit to their careers and are often seen as a waste of time to document.

2.3 Systemic Challenge 3: Collaboration

In computer software, open-source projects have revolutionized development by making code publicly accessible and encouraging global contributions, enabling developers to create better software collaboratively. However, the trajectory of the scientific community has moved in the opposite direction.

(Issac Newton, letter to Robert Hooke)

During the early scientific era, such as the 17th century, scientists prioritized sharing knowledge under natural philosophy and demonstrated open and collaborative attitudes, distancing themselves from rigid authorities. For example, despite their rivalry, Isaac Newton and Robert Hooke exchanged letters to share and critique each other's work, advancing knowledge collectively.

In contrast, modern science has become much more siloed. Researchers are driven by competition to secure funding and publish in journals with higher impact factors. Unpublished research is often kept confidential, and external sharing is strongly discouraged. Consequently, research labs within the same field naturally view one another as competitors, with few avenues to learn about each other's ongoing work.

Since most research builds incrementally on previous publications, there is a high likelihood that competing labs are conducting very similar studies. In the absence of shared research processes, parallel research on identical topics occurs simultaneously in multiple labs. This creates a highly inefficient and winner-takes-all environment where the lab that publishes results receives all the credit first. It is not uncommon for researchers to find that a similar study has been published just as they were about to complete their work, rendering much of their effort futile.

In the worst-case scenario, even within the same lab, students may withhold experimental materials or research findings from one another, competing internally rather than collaborating. As open-source culture has become a cornerstone of computer science, the modern scientific community must adopt a more open and collaborative culture to serve the greater public good.

3. How to Fix TradSci?

3.1 Many Have Already Tried

Researchers are well aware of these issues in the scientific community. While they recognize the problems, these challenges are deeply rooted structural issues that individuals cannot easily resolve. Nevertheless, numerous attempts have been made to address these problems over the years.

3.1.1 Fixing Centralized Funding

• Fast Grants: During the COVID-19 pandemic, Patrick Collison, the CEO of Stripe, identified inefficiencies in traditional funding processes and launched the Fast Grants program, raising $50M to support hundreds of projects. Grant decisions were made within 14 days, with funding amounts ranging from $10k to $500k—sizable amounts for researchers.

• Renaissance Philanthropy: Founded by Tom Kalil, a former science and technology policy advisor under Presidents Clinton and Obama, this nonprofit advisory organization connects donors with high-impact science and technology initiatives. Supported by Eric and Wendy Schmidt, it resembles the patronage system once prevalent among European scientists.

• hhmi: The Howard Hughes Medical Institute employs a unique funding model by supporting individual researchers rather than specific projects. By providing long-term funding, it alleviates the pressure for short-term results and allows researchers to focus on sustained investigations.

• experiment.com: This online crowdfunding platform enables researchers to introduce their work to the public and raise necessary funds from individual contributors.

3.1.2 Fixing Academic Journals

• PLOS ONE: PLOS ONE is an open-access scientific journal where anyone can freely read, download, and share articles. It evaluates papers based on scientific validity rather than impact and is well-known for publishing negative, null, or inconclusive results. Its streamlined publishing process helps researchers quickly disseminate findings. However, PLOS ONE charges researchers $1k–5k in article processing fees.

• arXiv, bioRxiv, medRxiv, PsyArXiv, SocArXiv: These are preprint servers allowing researchers to share drafts of their papers before formal publication in journals. They enable rapid dissemination of research findings, claim priority over specific topics, and offer opportunities for community feedback and collaboration, all while providing free access to papers for readers.

• Sci-hub: Founded by Kazakh computer programmer Alexandra Asanovna Elbakyan, Sci-hub provides free access to paywalled papers. Although illegal in most jurisdictions and subject to lawsuits from publishers like Elsevier, it has garnered praise for promoting free access to academic content while being criticized for violating the law.

3.1.3 Fixing Collaboration

• ResearchGate: A professional networking platform for researchers to share papers, ask and answer questions, and find collaborators.

• CERN: CERN, a nonprofit organization for particle physics research, conducts large-scale experiments that are difficult for individual labs to perform. It brings together researchers from multiple countries, with funding contributions based on participating nations’ GDPs.

3.2 DeSci, the New Wave

While the above efforts have made some progress in addressing the challenges of modern science, they have not created the transformative impact necessary to revolutionize the field. Recently, with the rise of blockchain technology, a new concept called Decentralized Science (DeSci) has gained attention as a potential solution to these structural issues. But what exactly is DeSci, and can it truly revolutionize the modern scientific ecosystem?

4. Enter DeSci

4.1 DeSci Overview

DeSci, short for Decentralized Science, refers to efforts to make scientific knowledge a public good by improving funding, research, peer review, and sharing of research outcomes within the scientific community. It strives for a system that is more efficient, fair, transparent, and accessible to everyone. Blockchain technology plays a central role in achieving these goals by leveraging the following features:

• Transparency: Except in privacy networks, blockchain networks are inherently transparent, allowing anyone to view transactions. This characteristic can enhance the transparency of project funding and peer review processes.

• Ownership: Blockchain assets are protected by private keys, making it easy to claim ownership. This feature enables researchers to monetize their data or assert intellectual property (IP) rights for funded research.

• Incentive Scheme: Incentives are at the core of blockchain networks. To encourage collaboration and active engagement, token incentives can be used to reward participants in various research processes.

• Smart Contracts: Deployed on neutral networks, smart contracts execute actions as defined in their code. They can be used to establish and automate interaction logic among participants transparently.

4.2 Potential DeSci Applications

As the name suggests, DeSci can be applied to various aspects of scientific research. ResearchHub categorizes potential applications of DeSci into the following five areas:

1. Research DAOs: These are decentralized autonomous organizations focused on specific research topics. Using blockchain technology, they manage research planning, funding, governance voting, and project administration transparently.

2. Publishing: Blockchain can decentralize and revolutionize the publishing process. Research papers, data, and code can be permanently recorded on the blockchain to ensure trustworthiness, allow free access to all, and incentivize peer reviewers with tokens, among other improvements.

3. Funding & IP: Researchers can easily secure funding from a global audience through blockchain networks. Additionally, by tokenizing research projects, token holders can participate in decision-making regarding the project's direction or share in future IP revenue.

4. Data: Blockchain enables secure, transparent storage, management, and research data sharing.

5. Infrastructure: This includes governance tools, storage solutions, community platforms, and identity systems that can be readily integrated into DeSci projects.

The best way to understand DeSci is to explore its ecosystem projects and examine how they address structural issues in modern science. Let’s look closely at some of the prominent projects within the DeSci ecosystem.

5. DeSci Ecosystem

(Source: ResearchHub)

5.1 Why the Ethereum Ecosystem Is Ideal for DeSci

Unlike applications in DeFi, gaming, or AI, DeSci projects are predominantly concentrated within the Ethereum ecosystem. This trend can be attributed to the following reasons:

• Credible Neutrality: Ethereum is the most neutral network among smart contract platforms. Given the nature of DeSci, where significant funding flows (e.g., research funding) are involved, values like decentralization, fairness, censorship resistance, and trustworthiness are essential. This makes Ethereum the optimal network for building DeSci projects.

• Network Effect: Ethereum boasts the largest user base and liquidity among smart contract networks. DeSci, a relatively niche sector compared to other applications, risks fragmentation if projects are spread across multiple networks. Such fragmentation could hinder project management due to liquidity and ecosystem-related challenges. Most DeSci projects are built on its network to leverage Ethereum's strong network effect.

• DeSci Infrastructure: Few DeSci projects are built entirely from scratch. Instead, many utilize existing frameworks like Molecule to accelerate development. Since most DeSci infrastructure tools are Ethereum-based, most projects in this space also operate on Ethereum.

For these reasons, the DeSci projects introduced in this discussion predominantly belong to the Ethereum ecosystem. Let’s now explore some representative projects within each sector of DeSci.

5.2 Funding & IP

5.2.1 Molecule

(Source: Molecule)

Molecule is a funding and tokenization platform for biopharma intellectual property. Researchers can secure funding from numerous individuals through blockchain, tokenize the project's IP, and funders can claim IP Tokens proportional to their contributions.

Catalyst, Molecule's decentralized fundraising platform, connects researchers and funders. Researchers prepare necessary documentation and project plans to propose their projects on the platform. Funders review these proposals and provide ETH to the projects they support. Once funding is completed, IP-NFTs and IP Tokens are issued, which funders can then claim.

(Source: Molecule)

An IP NFT represents a tokenized version of project IP on-chain, combining two legal agreements into a smart contract. The first legal agreement is the Research Agreement, signed between researchers and funders. It includes clauses on research scope, deliverables, schedule, budget, confidentiality, IP and data ownership, publication, results disclosure, licensing, and patent conditions. The second legal agreement is the Assignment Agreement, which transfers the Research Agreement to the IP NFT owner, ensuring that rights held by the current IP NFT owner can be transferred to a new owner.

IP Tokens represent fractional governance rights over the IP. Token holders can participate in key research decisions and access exclusive information. Although IP Tokens do not guarantee revenue sharing from the research, depending on the IP owner, profits from future commercialization may be distributed to IP Token holders.

(Source: Molecule)

The price of IP Tokens is determined by the Catalyst Bonding Curve, which reflects the relationship between the token’s supply and price. As more tokens are issued, their price increases. This incentivizes early contributions by allowing early funders to acquire tokens at a lower cost.

Here are some examples of successful funding cases through Molecule:

• Fang Laboratory at the University of Oslo: Fang Lab researches aging and Alzheimer’s disease. The lab is supported by VitaDAO through Molecule’s IP-NFT framework to identify and characterize new drug candidates for mitophagy activation, which positively impacts Alzheimer’s disease research.

• Artan Bio: Artan Bio focuses on tRNA-related research. It received $91,300 in funding from the VitaDAO community through Molecule’s IP-NFT framework.

5.2.2 Bio.xyz

(Source: Bio.xyz)

Bio.xyz is a curation and liquidity protocol for DeSci that is comparable to an incubator supporting BioDAOs. The goals of Bio.xyz are:

• Curation, creation, and acceleration of new BioDAOs funding science on-chain.

• Perpetual funding and liquidity for BioDAOs and on-chain biotech assets.

• Standardization of BioDAO frameworks, tokenomics, and data/product suites.

• Generation and commercialization of scientific IP and data.

BIO token holders vote on which new BioDAOs will join the ecosystem. Once a BioDAO is approved to join the BIO ecosystem, token holders who voted for it can participate in the initial private token auction. This process resembles a whitelisted pre-seed round.

The governance tokens of the approved BioDAO are paired with BIO tokens and added to a liquidity pool, eliminating the need for BioDAOs to worry about liquidity for their governance tokens (e.g., VITA/BIO). Additionally, Bio.xyz runs the bio/acc rewards program, providing BIO token incentives to BioDAOs as they achieve key milestones.

That’s not all. BIO tokens act as a meta-governance token for multiple BioDAOs within the ecosystem. This enables BIO holders to participate in the governance of various BioDAOs. Furthermore, the BIO network provides incubated BioDAOs with a $100,000 grant and acquires 6.9% of the BioDAO’s token supply for the treasury. This increases the protocol's AUM (Assets Under Management) and accrues value to BIO tokens.

Bio.xyz leverages Molecule’s IP NFT and IP Tokens framework for managing and owning IP. For instance, VitaDAO has successfully issued IP Tokens such as VitaRNA and VITA-FAST within the Bio ecosystem. Below is a list of Research DAOs currently being incubated through Bio.xyz, which will be discussed in detail in the next section:

• Cerebrum DAO: Focuses on preventing the onset of neurodegeneration.

• PsyDAO: Dedicates to conscious evolution through safe, accessible psychedelic experiences.

• cryoDAO: Contributes to cryopreservation research projects.

• AthenaDAO: Works to advance women’s health research.

• ValleyDAO: Supports synthetic biology research.

• HairDAO: Collaborates to develop new treatments for hair loss.

• VitaDAO: Focuses on human lifespan.

In summary, Bio.xyz curates BioDAOs and provides token frameworks, liquidity services, grants, and incubation support. When the IPs of BioDAOs within the ecosystem successfully commercialize, the value of Bio.xyz's treasury increases, creating a virtuous cycle.

5.3 Research DAOs

5.3.1 VitaDAO

Regarding the most well-known Research DAO, VitaDAO often comes to mind first. Its fame stems from being an early DeSci project and receiving lead investment from Pfizer Ventures in 2023. VitaDAO funds projects focused on longevity and aging research, having supported over 24 projects with more than $4.2M in funding. In return for funding, VitaDAO acquires IP NFTs or equity in companies, utilizing Molecule.xyz's framework for the IP NFTs.

VitaDAO leverages blockchain transparency by making its treasury publicly accessible. The treasury’s value amounts to approximately $44M, including around $2.3M in equities and $29M in tokenized IP, among other assets. VITA token holders participate in governance votes to shape the DAO’s direction and gain access to various healthcare services.

The most notable projects funded by VitaDAO are VitaRNA and VITA-FAST. Both projects’ IPs have been tokenized and are actively traded, with the market cap of VITARNA at approximately $13M and VITA-FAST at $24M. Both projects hold regular calls with VitaDAO to update their progress.

• VitaRNA: VitaRNA is an IP Token project led by the biotechnology company Artan Bio. Successfully funded in June 2023, the project issued an IP NFT, fractionalized into IP Tokens in January 2024. Their innovative research focuses on suppressing arginine nonsense mutations, particularly the CGA codon, which is critical in proteins associated with DNA damage, neurodegeneration, and tumor suppression.

• VITA-FAST: VITA-FAST is an IP Token project from the Viktor Korolchuk Lab at Newcastle University. The project focuses on discovering novel autophagy activators. Autophagy, a cellular process whose decline contributes to biological aging, is stimulated to explore therapeutic approaches that combat aging and related diseases, ultimately aiming to enhance human healthspan.

5.3.2 HairDAO

HairDAO is an open-source R&D network where patients and researchers collaborate to develop treatments for hair loss. According to Scandinavian Biolabs, hair loss affects 85% of men and 50% of women in their lifetime. However, only treatments like Minoxidil, Finasteride, and Dutasteride exist on the market. Notably, Minoxidil was FDA-approved in 1988 and Finasteride in 1997.

Even these approved treatments provide limited effects, such as slowing or temporarily halting hair loss, rather than offering a cure. Developing hair loss treatments is slow for several reasons:

• Complex causes: Hair loss is caused by multiple factors, including genetics, hormonal changes, and immune responses, making it challenging to develop effective, targeted treatments.

• High development costs: Drug development requires substantial time and investment, but because hair loss is non-life-threatening, it often ranks lower in research funding priorities.

HairDAO rewards patients with HAIR governance tokens for sharing their treatment experiences and data through the app. Holders of HAIR tokens can participate in DAO governance votes, enjoy discounts on HairDAO shampoo products, and stake tokens to access confidential research data faster.

5.3.3 Others

• CryoDAO: CryoDAO focuses on cryopreservation research, with a treasury exceeding $7M and funding for five projects. CRYO token holders can participate in governance votes and may receive early or exclusive access to breakthroughs and data from funded research.

• ValleyDAO: ValleyDAO aims to address climate challenges by funding synthetic biology research. Synthetic biology seeks to sustainably synthesize nutrients, fuels, and medicines using biological organisms, a critical technology for combating climate change. ValleyDAO has funded several projects, including research by Prof. Rodrigo Ledesma-Amaro at Imperial College London.

• CerebrumDAO: CerebrumDAO focuses on brain health research, particularly Alzheimer’s prevention. Its Snapshot page showcases numerous proposals from projects seeking funding. Decision-making is decentralized and conducted through DAO member votes.

5.4 Publishing

5.4.1 ResearchHub

(Source: ResearchHub)

ResearchHub is the leading DeSci publishing platform, aiming to become the "GitHub for science." Founded by Coinbase CEO Brian Armstrong and Patrick Joyce, ResearchHub successfully raised $5M in a Series A round in June 2023, led by Open Source Software Capital.

ResearchHub is a tool for open publication and discussion of scientific research, incentivizing researchers to publish, peer review, and curate through its native RSC tokens. Its key features include:

Grants

(Source: ResearchHub)

Using RSC tokens, users can create grants to request specific tasks from other ResearchHub users. Grant types include:

• Peer Review: Request reviews for manuscripts.

• Answer to Question: Request answers to specific questions.

Funding

(Source: ResearchHub)

In the Funding tab, researchers can upload research proposals and receive funding from users in RSC tokens.

Journals

(Source: ResearchHub)

The Journals section archives papers from peer-reviewed journals and preprint servers. Users can browse literature and engage in discussions. However, many peer-reviewed papers are behind paywalls, and users can only access summaries written by others.

Hubs

(Source: ResearchHub)

Hubs archive preprint papers categorized by field. This section contains all papers in open-access, allowing anyone to read the full content and engage in discussions.

Lab Notebook

The Lab Notebook is a collaborative online workspace where multiple users can co-author papers. Like Google Docs or Notion, this feature allows seamless publishing directly on ResearchHub.

RH Journal

(Source: ResearchHub)

RH Journal is ResearchHub's in-house journal. It boasts an efficient peer-review process that is completed within 14 days and decisions made within 21 days. Additionally, it incorporates an incentive system for peer reviewers, addressing the misaligned incentive issues common in traditional peer-review systems.

RSC Token

(Source: ResearchHub)

RSC tokens are ERC-20 tokens used within the ResearchHub ecosystem, with a total supply of 1 billion. RSC tokens drive engagement and support ResearchHub’s vision of becoming a fully decentralized open platform. Their utilities include:

• Governance voting

• Tipping other users

• Bounty programs

• Incentives for peer reviewers

• Rewards for curating research papers

5.4.2 ScieNFT

ScieNFT is a decentralized preprint server where researchers can publish their work as NFTs. The format of the publication can range from simple figures and ideas to datasets, artistic works, methods, and even negative results. Preprint data is stored using decentralized storage solutions like IPFS and Filecoin, while NFTs are uploaded to the Avalanche C-Chain.

While using NFTs to identify and track ownership of work is an advantage, a notable drawback is the unclear benefits of purchasing these NFTs. Additionally, the marketplace lacks effective curation.

5.4.3 deScier

(Source: deScier)

deScier is a decentralized science journal platform. Like publishers like Elsevier or Springer Nature, which manage multiple journals under their umbrella, deScier also hosts various journals. Copyright for all papers remains 100% with the researchers, and peer review is part of the process. However, as noted below, a significant limitation is the low number of papers published in the journals and the slow rate of uploads.

5.5 Data

5.5.1 Data Lake

Data Lake’s software enables researchers to integrate various user recruitment channels, track their effectiveness, manage consents, and conduct preselection surveys while giving users control over their data. Researchers can share and easily manage patient consent for data use among third parties. Data Lake uses the Data Lake Chain, an L3 network based on Arbitrum Orbit, to manage patient consent.

5.5.2 Welshare Health

(Source: Welshare Health)

In traditional medical research, the most significant bottlenecks are delays in recruiting clinical trial participants and the lack of patients. Additionally, while patient medical data is valuable, it poses risks of misuse. Welshare aims to address these challenges using Web3 technology.

Patients can securely manage their data, monetize it to earn income, and access personalized healthcare services. Conversely, medical researchers benefit from easier access to diverse datasets, facilitating their research.

Through a Base Network-based application, users can selectively provide data to earn in-app reward points, which can later be converted into crypto or fiat currency.

5.5.3 Hippocrat

Hippocrat is a decentralized healthcare data protocol that allows individuals to securely manage their health data using blockchain and zero-knowledge proof (ZKP) technology. Its first product, HippoDoc, is a telemedicine application that provides healthcare consultations using a medical database, AI technology, and assistance from healthcare professionals. Throughout this process, patient data is securely stored on the blockchain.

5.6 DeSci Infrastructure

5.6.1 Ceramic

Ceramic is a decentralized event streaming protocol that allows developers to create decentralized databases, distributed compute pipelines, authenticated data feeds, and more. These features make it well-suited for DeSci projects, enabling them to utilize Ceramic as a decentralized database:

• Data on the Ceramic network is permissionlessly accessible, allowing researchers to share and collaborate on data.

• Actions such as research papers, citations, and reviews on the Ceramic network are represented as “Ceramic streams.” Individual streams can only be modified by their originating author account, ensuring IP provenance.

• Ceramic also provides infrastructure for verifiable claims, enabling DeSci projects to adopt its reputation infrastructure.

5.6.2 bloXberg

bloXberg is a blockchain infrastructure established under the leadership of the Max Planck Digital Library in Germany, with participation from renowned research institutions such as ETH Zurich, Ludwig Maximilian University of Munich, and the IT University of Copenhagen.

bloXberg is designed to innovate various processes in scientific research, such as research data management, peer review, and intellectual property protection. Utilizing blockchain decentralizes these processes, enhancing transparency and efficiency in research. Researchers can securely share and collaborate on research data using the blockchain.

6. Is DeSci Really a Silver Bullet?

We’ve explored the structural issues in modern science and how DeSci aims to address them. But hold on a second. Can DeSci truly revolutionize the scientific community and play a central role, as the crypto community claims? I don’t believe so. However, I do think DeSci has the potential to play a supportive role in certain areas.

6.1 What Blockchain Can and Cannot Fix

Blockchain isn’t magic. It cannot solve every problem. We must clearly distinguish what blockchain can address and cannot.

6.1.1 Funding

DeSci is expected to excel in funding scenarios that meet the following conditions:

• Small-scale grants

• Research with potential for commercialization

The scale of funding in the scientific community varies widely, ranging from tens of thousands to millions or even tens of millions of dollars. For large-scale projects requiring significant capital, centralized funding from governments or corporations is inevitable. However, smaller projects can feasibly secure funding through DeSci platforms.

From the perspective of researchers conducting small-scale projects, the burden of extensive paperwork and lengthy funding review processes can be overwhelming. In this context, DeSci funding platforms, which provide funding quickly and efficiently, are highly appealing.

That said, to increase the likelihood of a research project receiving funding from the public through a DeSci platform, there must be a reasonable prospect of commercialization, such as through patents or technology transfer. This provides an incentive for the public to invest in the project. However, most modern scientific research is not geared toward commercialization but is instead supported to enhance national or corporate technological competitiveness.

In summary, fields well-suited for funding on DeSci platforms include biotech, healthcare, and pharmaceuticals. The focus of most current DeSci projects on these areas aligns with this reasoning. These fields have a high likelihood of commercialization if the research succeeds. Moreover, while significant funding is required for eventual commercialization, the initial phases of research typically demand less funding than other fields, making DeSci platforms a favorable option for raising capital.

I question whether DeSci can enable long-term research. While a small number of researchers might be supported by altruistic and voluntary funders to pursue long-term studies, this culture is unlikely to spread widely throughout the scientific community. Even with DeSci platforms leveraging blockchain, no inherent causal link suggests they can sustain long-term funding. If one were to seek a connection between blockchain and long-term research deliberately, one possible consideration could be milestone-based funding through smart contracts.

6.1.2 Journals

Ideally, the area where DeSci could bring the most innovation is academic journals. Through smart contracts and token incentives, DeSci can potentially restructure the profit model dominated by journals into one centered around researchers. However, in reality, this will be challenging.

The most critical factor for researchers building their careers is publishing papers. In academia, a researcher’s capabilities are primarily judged by the journals they publish in, their citation counts, and their h-index. Human nature inherently leans on authority—a fact unchanged from prehistoric times to the present. For example, an unknown researcher can become a star overnight by publishing in top-tier journals like Nature, Science, or Cell.

While qualitative evaluations of researchers' skills would be ideal, such evaluations rely heavily on peer references, making quantitative assessments almost unavoidable. Because of this, journals hold immense power. Despite monopolizing the profit model, researchers have no choice but to comply. For DeSci journals to gain more influence, they must build authority, but achieving the reputation that traditional journals have accumulated over a century through token incentives alone is highly challenging.

While DeSci may not completely transform the journal landscape, it can undoubtedly contribute to specific areas, such as peer review and negative results.

As previously mentioned, peer reviewers currently receive little to no incentives, which lowers the quality and efficiency of peer reviews. Providing token incentives to reviewers could improve review quality and elevate the journal’s standards.

Additionally, token incentives could bootstrap a journal network dedicated solely to publishing negative results. Since reputation is less critical for journals exclusively publishing negative results, the combination of token rewards would incentivize researchers to publish their findings in such journals.

6.1.3 Collaboration

In my view, blockchain is unlikely to address the fierce competition in modern science significantly. Unlike in the past, the number of researchers today is much greater, and every achievement directly impacts career progression, making competition inevitable. It is unrealistic to expect blockchain to resolve overall collaboration challenges in the scientific community.

On the other hand, within small groups like research DAOs, blockchain can effectively promote collaboration. Researchers in DAOs align incentives through tokens, share a common vision, and log achievements on the blockchain via timestamps to gain recognition. I hope to see an increase in the number and activity of research DAOs not just in the biotech field but across other disciplines.

7. Final Thoughts: DeSci Needs a Bitcoin Moment

The modern scientific community faces numerous structural challenges, and DeSci offers a compelling narrative for addressing them. While DeSci may not revolutionize the entire scientific ecosystem, it can gradually expand through researchers and users who find value in it. Eventually, we may see a balance between TradSci and DeSci. Just as Bitcoin, once dismissed as a toy for computer geeks, now has major traditional financial institutions entering the market, I hope DeSci will similarly gain long-term recognition and achieve its "Bitcoin moment."

1. Everyone Wants it, Few can Wield it

1.1 The Land of Opportunity: Bitcoin

(Source: companiesmarketcap)

Bitcoin, created in 2008 by an anonymous developer, has grown into a massive asset, ranking 7th in market capitalization among all asset classes. It is now recognized not only by financial institutions but even by the U.S. President. Currently, Bitcoin’s market capitalization is similar to that of silver. Considering that Bitcoin's adoption is still relatively low and that its market capitalization is only one-tenth that of gold, its future growth potential remains highly promising.

Despite Bitcoin’s immense growth as an asset, there is still a significant shortcoming—its level of utilization. Traditional assets like stocks and bonds can be used in a wide range of financial products, but Bitcoin’s financial applications are still very limited, both technically and practically. Similar to the frontier days of the American West, Bitcoin represents an untapped land of opportunity.

1.2 Attempts to Utilize BTC

Due to its enormous market capitalization, numerous companies and protocols have sought to leverage Bitcoin for additional credit creation. The main attempts to utilize BTC so far include:

• Centralized Finance: Traditional financial institutions offer various Bitcoin-based financial products. CME provides Bitcoin futures and options, Coinbase offers BTC-backed loans, and multiple institutions have launched BTC-based ETFs for investors.

• Custodian Bridging: Services like WBTC and cbBTC wrap BTC in a centralized manner, allowing it to be used on other networks. By depositing BTC with custodians like BitGo or Coinbase, users receive an equivalent amount of WBTC or cbBTC issued on the Ethereum network.

• On-chain Bridging: To eliminate the reliance on centralized custodians, various protocols have attempted to securely bridge BTC to other networks. However, achieving a completely trustless BTC bridging mechanism remains highly challenging, as some level of trust assumptions is inevitable.

• Scaling Solutions: Efforts to use BTC on sidechains and Bitcoin L2 solutions have increased recently. However, these approaches still involve additional trust assumptions. The Taproot Wizards team is working on mitigating this issue using OP_CAT.

• BTC-based Stablecoins: Protocols like Yala and Avalon have emerged, issuing stablecoins backed by BTC in a manner similar to MakerDAO. However, these solutions also face the fundamental issue of requiring trust assumptions when bridging BTC.

Examining these attempts to utilize BTC reveals a common challenge—it is difficult to use Bitcoin in a native manner. One of Bitcoin’s greatest strengths is its security, but if additional trust assumptions weaken this security, it creates a significant entry barrier for BTC holders. This is the primary reason why Bitcoin’s level of utilization remains relatively low.

1.3 Babylon: Native Utilization of BTC

This is where Babylon comes into focus. Babylon enables BTC holders to stake their Bitcoin natively on the Bitcoin network and participate in validating other PoS protocols, earning additional rewards.

Thanks to the advantage of utilizing BTC without additional trust assumptions, Babylon has rapidly achieved over $5 billion in TVL. The TVL could have been even higher if there were no staking limits on BTC.

But wait, Bitcoin’s scripting language is non-Turing complete, meaning it cannot easily support complex smart contracts. So, how does Babylon manage to achieve this functionality? In this article, we will explore the specific mechanisms behind Babylon’s operation.

2. Babylon

Like building the Tower of Babel, can we ever reach true native BTC utilization?

2.1 Babylon Overview

Babylon’s mission is scaling Bitcoin to secure the decentralized world. While it is widely known as a BTC staking protocol, Babylon also offers Bitcoin timestamping services, forming a suite of BTC security-sharing protocols.

Babylon is composed of two main protocols:

• Bitcoin Timestamping: This allows PoS chains to checkpoint their block data onto the Bitcoin network. By doing so, PoS chains can mitigate long-range attacks, reduce stake unbonding periods, protect critical transactions, and benefit from Bitcoin’s censorship resistance at the network level.

• Bitcoin Staking: This enables BTC holders to freeze their BTC natively on the Bitcoin network and participate in the validation of other PoS protocols, earning additional rewards in the process.

2.2 Babylon Architecture

(Source: Babylon)

The fundamental architecture of Babylon is illustrated in the diagram above, with the Babylon Chain, built on the Cosmos SDK, at its core. In addition to the Babylon Chain, several peripheral programs facilitate BTC staking and communication with Bitcoin and other Consumer Zones. Consumer Zones refer to PoS chains that record checkpoints on the Bitcoin network through Babylon.

The Babylon Chain consists of various modules that perform essential functions within the ecosystem, including managing the validator set, tracking Bitcoin block headers, submitting checkpoints to the Bitcoin network, and managing the active finality provider set related to BTC staking. For reference, a finality provider is similar to an AVS operator in EigenLayer, meaning it participates in validating other PoS protocols.

Additionally, Babylon has implemented several supporting programs to facilitate smooth communication between the Bitcoin network and the Babylon Chain:

• Vigilantes: A suite of programs that acts as a data relayer between Bitcoin and Babylon.

• Monitors: A suite of programs that ensures the consistency between the Babylon Chain and the Bitcoin network.

Through this ecosystem, Babylon enables the crypto industry to leverage Bitcoin’s strong security and deep liquidity. Now, let’s explore Babylon’s two core features in more detail: Bitcoin Timestamping and Bitcoin Staking.

3. How Bitcoin Timestamping Works

3.1 Why is Stake Unbonding Slow?

Anyone who has staked tokens before would know that unstaking typically requires a waiting period of 1 to 2 weeks. During this time, the tokens cannot be used or earn interest, causing inefficiencies. But why does unstaking require a waiting period? Why not allow instant withdrawals?

The simplest reason is network security. If unstaking were instant, large amounts of tokens could be unstaked in response to market fluctuations, significantly weakening network security. However, beyond security, there is another fundamental reason: to prevent long-range attacks.

3.2 Long-Range Attack

(Source: AP)

A long-range attack refers to an attack in which malicious validators create a new fork starting from past blocks, attempting to replace the current canonical chain. If the malicious forked chain becomes as long as or longer than the canonical chain, newly joining nodes in the network may become confused about which chain is the legitimate one, leading to potential issues. But wait—is this even possible?

In a PoW network, a long-range attack is nearly impossible. To catch up with the current canonical chain, attackers would need to recreate new blocks from the past while exceeding the computing power of the existing network, which is practically infeasible.

Similarly, in a properly functioning PoS network, this attack is also impossible. Creating a new fork would require malicious validators to sign multiple conflicting blocks, which is considered double-signing—a violation of the protocol that results in slashing penalties.

However, what if unstaking were allowed immediately?

Unlike PoW networks, PoS networks do not require massive computational power to generate blocks. This means that if malicious validators unstake their assets from the existing chain and then create a new forked chain from a past block where their validator keys were still valid, they could potentially catch up with the current canonical chain. In this scenario, newly joining nodes in the network might struggle to determine which chain is the legitimate one, leading to potential confusion and security risks.

(Source: Babylon)

If a long-range attack succeeds, malicious validators can exploit bridging mechanisms to steal funds. For example, suppose a malicious attacker named John transfers 1M RUG tokens from the RugPull chain to Osmosis and exchanges them for OSMO tokens. This transfer happens through IBC, which works by locking the original RUG tokens on the RugPull chain while minting an equivalent amount of RUG tokens on the Osmosis chain.

(Source: Babylon)

If we assume that John successfully executes a long-range attack on the RugPull chain, he can maliciously omit the transaction that locked RUG tokens to send them to the Osmosis chain in the new forked chain. As a result, John would effectively acquire OSMO tokens for free.

To prevent long-range attacks, a stake unbonding period of a certain length is necessary. Malicious actors cannot execute a long-range attack during the unbonding period (if they attempt it, they will face slashing penalties). Additionally, during this period, network participants can reach a social consensus regarding which chain is the canonical chain. As a result, even if a long-range attack occurs later, the malicious forked chain is unlikely to be accepted by the network.

3.3 Let’s Reduce Stake Unbonding Time with BTC Timestamping

The stake unbonding period is an effective method for preventing long-range attacks, but it comes with some drawbacks.

The first issue is that it relies on social consensus to counteract attacks. While off-chain communication among participants over a long enough period can play a crucial role, it is not a 100% foolproof solution.

The second issue is that, as mentioned earlier, a longer unstaking period negatively affects user experience and liquidity.

Babylon introduces a solution called Bitcoin Timestamping, which enables PoS chains to significantly reduce unstaking periods to just a few hours. This allows PoS chains to record canonical chain block data onto the Bitcoin network.

With timestamping, even if malicious validators attempt a long-range attack and claim their forked chain is the canonical chain, their attack will not succeed—because the original canonical chain data is already securely recorded on the Bitcoin network. As long as Bitcoin’s security remains intact, the attack is guaranteed to fail. Since this approach eliminates the need for social consensus, it allows for a drastic reduction in the required unbonding period.

(Source: Babylon)

Here, Bitcoin Timestamping is recorded using the OP_RETURN opcode in the Bitcoin network. OP_RETURN is an instruction that allows storing up to 80 bytes of arbitrary data on the Bitcoin network. Unlike regular Bitcoin transactions, OP_RETURN cannot be used for fund transfers and does not generate UTXOs.

One key consideration is whether all PoS chains can directly create checkpoints on the Bitcoin network. Bitcoin blocks are small in size, have a block time of 10 minutes, and OP_RETURN can only store a maximum of 80 bytes of data. If numerous PoS chains were to send frequent checkpointing transactions, the Bitcoin network would not be able to handle the load.

To solve this issue, Babylon introduces the Babylon Chain, which aggregates checkpoints from multiple PoS chains via IBC and then submits a single aggregated checkpoint to the Bitcoin network.

A key component of this process is the Vigilante Relayer, an entity responsible for reading checkpoints from a Babylon node, packaging them into OP_RETURN transactions, and then submitting them to the Bitcoin network. The system requires at least one honest and live Vigilante Relayer to function properly.

(Source: Babylon)

BTC timestamping occurs as follows: PoS chains submit checkpoints containing block information to the Babylon chain. The Babylon chain then submits a checkpoint of Babylon blocks to the Bitcoin network at the final block of each epoch.

(Source: Babylon)

Even if a long-range attack occurs, the malicious forked chain’s checkpoint will always have a timestamp later than the canonical chain’s checkpoint. This means that network participants can simply check the Bitcoin network’s checkpoint to easily identify malicious forks. Since this approach eliminates the need for social consensus, the stake unbonding period can be reduced from several weeks to just a few hours.

3.4 More Than Fast Stake Unbonding

Babylon’s Bitcoin Timestamping does more than just improve UX and liquidity efficiency by reducing PoS chains’ unstaking periods—it also provides various additional benefits.

3.4.1 Slow Finality for Important Transactions

By adopting slow finality through Babylon, PoS chains can achieve a level of security comparable to Bitcoin. When a PoS block containing a specific transaction is timestamped on the Bitcoin network and confirmed by six Bitcoin blocks, the transaction becomes irreversible—as long as Bitcoin’s security remains intact.

This mechanism is useful for processing high-value transactions, such as real estate purchases, where absolute security is necessary. Additionally, for newly launched Cosmos zones, which may have weaker security, implementing slow finality can provide an extra layer of protection for safe transaction processing.

3.4.2 Bitcoin-Level Censorship Resistance

Bitcoin Timestamping can also help restore liveness in the event of a censorship attack on a PoS chain. To address this, Babylon introduces a special concept called rollup mode.

In a traditional PoS chain, at least two-thirds (2/3) of validators must be honest to maintain censorship resistance. However, with Babylon’s rollup mode, only one-half (1/2) of validators need to be honest to achieve censorship resistance, significantly improving the chain’s resilience against attacks.

(Source: Babylon)

If a PoS chain user believes that a specific transaction is being censored, they can submit a censorship complaint (the red section in the diagram) to the Babylon chain, initiating the process of entering rollup mode. The censorship complaint contains the hash of the censored transaction.

If, after six Bitcoin block confirmations, the suspected censored transaction still hasn’t been included in the PoS chain, honest validators will submit their views of the PoS chain to Babylon. If, after an additional six Bitcoin block confirmations, no checkpoint related to the censored transaction is detected in any Bitcoin block, honest validators and users will enter rollup mode.

In rollup mode, any validator can propose a bundle of PoS transactions, and if validators holding at least one-half (1/2) of the total stake sign the bundle, the transaction will be finalized on the Bitcoin network, effectively preventing censorship.

4. How Bitcoin Staking Works

4.1 Bitcoin Staking Overview

Bitcoin Timestamping allows PoS chains to leverage Bitcoin’s security to reduce stake unbonding periods and enhance censorship resistance, but this only partially utilizes Bitcoin’s security.

Beyond Bitcoin Timestamping, Babylon introduces Bitcoin Staking, which natively implements BTC staking using Bitcoin’s script language. This allows other PoS protocols to benefit from staked BTC’s cryptoeconomic security. The staking protocol is designed as a modular plug-in, making it easily adaptable for various PoS consensus protocols.

For BTC holders, Babylon’s Bitcoin Staking is an attractive investment opportunity since they can stake BTC at Bitcoin’s security level without relying on external entities, while also earning rewards from external protocols.

Let’s define some key terms:

• Protocols that leverage staked BTC’s cryptoeconomic security through Babylon’s Bitcoin Staking are called BSNs (Bitcoin Secured Networks)—analogous to EigenLayer’s AVS (Actively Validated Services) concept.

• Entities that receive delegated BTC from stakers and participate in validating BSNs are called Finality Providers—similar to EigenLayer’s AVS operators.

But wait—unlike Ethereum, the Bitcoin network is not Turing-complete, making it difficult to implement complex staking contracts. So how does Babylon achieve this?

Let’s explore the details with an example from Babylon’s blog.

4.2 How Staking Contract is Implemented

4.2.1 Locking

// Contract V0: adding a locking condition to Alice's staking UTXO

condition-1 (locking): time_lock = 1000 & alice_public_key

Let's assume that Alice stakes BTC and also acts as a Finality Provider. To implement BTC staking, a mechanism to lock BTC is required. This is achieved by setting one of the UTXO spending conditions so that only Alice (the BTC owner) can withdraw the funds after a certain time period (time_lock = 1000) using her alice_public_key.

4.2.2 Slashing

// Contract V1: adding naive slashing

condition-1 (locking): time_lock = 1000 & alice_public_key; OR

condition-2 (slashing): alice_eots_public_key

One of the essential components that must be implemented in staking is slashing. If a malicious action occurs, an incentive mechanism can be enforced by burning the staked BTC. To achieve this, a second UTXO spending condition is set so that slashing can occur if someone holds Alice's EOTS key.

Here, EOTS (Extractable One-Time Signature) is a signature implemented using Schnorr signatures, which was introduced after Bitcoin's Taproot upgrade. Simply put, it is an algorithm that ensures that if a malicious actor double-signs two different blocks at the same height using the same key, their secret key is publicly revealed.

Looking at this in more detail, a Schnorr signature consists of a private key x, a public key P=xG, and a random nonce k. The signing process is as follows: a random nonce k is generated, and the public value R=kG is computed from the nonce. Then, a hash value e is computed from the message M and R, and the signature value s is computed based on the nonce and e, where s = k + ex. The final Schnorr signature consists of (s, R).

The core idea of EOTS is that if the same key is used twice for signing, the private key is exposed. If Alice signs two different messages using the same nonce k, then the first signature is s1= k + e_1x and the second signature is s2= k + e_2x. Since s1, s2, e1, e2 are publicly known, anyone can solve for Alice’s private key x using the equation x=(s1 - s2)/(e1 - e2).

Using this mechanism, if Alice maliciously signs two different messages using the same EOTS key during the BSN validation process, anyone who detects this can extract Alice's EOTS secret key. Once the EOTS secret key is revealed, the attacker can either steal Alice's staked BTC or burn Alice's staked BTC as a penalty.

4.2.3 Enforcing Burning

// Contract V2

condition-1 (locking): time_lock = 1000 & alice_public_key; OR

condition-2 (slashing): alice_eots_public_key & covenant_committee_quorum

// Slashing transaction V0

inputs:
  - input-1: the staking UTXO, spent using condition-2 above

outputs:
  - output-1: value = 0.99 Bitcoin, owner = `0000...0000`

// Pre-approval V0: enforcing burning
//
// Covenant committee pre-signs the above slashing tx as its pre-approval

Since we have previously discussed the conditions under which slashing occurs, let's now examine how slashing is actually enforced. Enforcing slashing is crucial because, if Alice engages in malicious behavior, she might attempt to withdraw her BTC before someone detects the misconduct, extracts her EOTS secret key, and burns her BTC.

To prevent this, slashing must be implemented in a way that forcibly transfers the BTC to a pre-defined burn address (0000…0000). To achieve this, the second UTXO spending condition includes a Covenant Committee Quorum. The Covenant Committee is responsible for verifying whether slashing is legitimate. By incorporating a multi-signature (M-of-N) scheme, the system ensures that Alice cannot unilaterally withdraw her BTC to her own wallet before slashing is executed.

The advantage of this approach is that, as long as Alice behaves honestly, her EOTS signature is never exposed, meaning the Covenant Committee cannot seize her funds. Therefore, Alice does not need to trust the Covenant Committee, as they cannot act against her unless she engages in malicious behavior.

4.2.4 Safe Delegation

/ Contract V3: enabling safe delegation

condition-1 (locking): time_lock = 1000 & alice_public_key; OR

condition-2 (slashing): alice_public_key & validator_eots_public_key & covenant_committee_quorum

// Slashing transaction V0

inputs:
  - input-1: the staking UTXO, spent using condition-2 above

outputs:
  - output-1: value = 0.99 Bitcoin, owner = `0000...0000`

// Pre-approval V1
//
// Alice pre-signs the slashing tx as her pre-approval.
//
// Covenant committee pre-signs the slashing tx as its pre-approval.

Alice can directly stake BTC and participate in validating other PoS protocols as a finality provider. However, most users will choose to delegate their BTC staking.

To implement this, adding the validator's EOTS key to the second condition ensures that if the validator engages in malicious behavior, Alice’s BTC can be burned. However, the issue here is that if the validator colludes with the covenant committee, they could steal Alice's BTC, forcing Alice to trust the validator.

A simple solution to this problem is to also include Alice’s public key in the second condition. This way, burning BTC would require Alice’s signature as well, preventing unauthorized BTC theft.

To achieve this, Alice pre-signs a transaction stating that "if slashing occurs, BTC must be sent to a burn address." In this case, if the validator acts maliciously and their EOTS key is exposed, and if the covenant committee executes a multi-signature, the BTC will be sent to the burn address, enforcing the slashing process.

4.2.5 Preventing Malicious Attack with Enforcing Atomic Slashing

/ Contract V3

condition-1 (locking): time_lock = 1000 & alice_public_key; OR

condition-2 (slashing): alice_public_key & validator_eots_public_key & covenant_committee_quorum

// Slashing transaction V0

inputs:
  - input-1: the staking UTXO, spent using condition-2 above

outputs:
  - output-1: value = 0.99 Bitcoin, owner = `0000...0000`

// Pre-approval V2: enforcing atomic slashing when delegate
//
// Alice's pre-approval is an adaptor signature of the slashing tx
// she generated using the validator's EOTS public key.
//
// Covenant committee pre-signs the slashing tx as its pre-approval.

What if a malicious validator targets only specific stakers for slashing? To prevent this, Babylon introduces Adaptor Signatures.

Alice encrypts her signature using the validator’s EOTS public key as an adaptor signature. If the validator attempts to slash only Alice, they must use their EOTS private key. Due to the nature of Adaptor Signatures, this would result in the exposure of the validator’s EOTS private key, removing any incentive for validators to engage in malicious behavior.

4.2.6 Implementing Partial Slashing

// Contract V3

condition-1 (locking): time_lock = 1000 & alice_public_key; OR

condition-2 (slashing): alice_public_key & validator_eots_public_key & covenant_committee_quorum

// Slashing transaction V1: enabling partial slashing

inputs:
  - input-1: the staking UTXO, spent using condition-2 above

outputs:
  - output-1: value = 0.09 Bitcoin, owner = `0000...0000`
  - output-2: value = 0.9 Bitcoin,
              conditions:
              - condition-1: time_lock = 500 & alice_public_key

// Pre-approval V2
//
// Alice's pre-approval is an adaptor signature of the slashing tx
// she generated using the validator's EOTS public key.
//
// Covenant committee pre-signs the slashing tx as its pre-approval.

But don’t you think burning all the Bitcoin in the event of slashing is too extreme? To address this, only a portion of the Bitcoin (e.g., burning just 10% while returning the remaining 90% after a certain period) can be burned. This can be implemented by splitting the outputs of the slashing transaction into two, as described above.

4.2.7 Restaking More!

// Contract V4: Enabling restaking

condition-1 (locking): time_lock = 1000 & alice_public_key; OR

condition-2 (slashing): alice_public_key
                        & any signature from list [validator_eots_public_key]
                        & covenant_committee_quorum

Alice’s delegated BTC can participate in the validation of multiple PoS protocols, not just a single one. If a validator participates in the validation of different PoS protocols using the same EOTS key, any leakage in one place could affect the other systems. Therefore, Babylon’s finality providers must use different EOTS keys for different PoS systems, and a list of EOTS keys is introduced in the second condition.

4.3 Summary

Unlike PoS networks such as Ethereum or Solana, Bitcoin's network operates on PoW, so the concept of staking doesn’t inherently exist. However, Babylon has implemented BTC locking, slashing, and delegation features necessary for staking through the characteristics of UTXOs, Bitcoin's scripting language, and various signature algorithms. This allows BTC holders to earn additional profits natively by utilizing BTC, without needing bridges or custody services.

5. Unleashing BTC’s Potential to the Decentralized World

Aside from the Lightning Network, no other protocol fully inherits the security of the Bitcoin network. However, just like the Bitcoin network, the functionality of the Lightning Network is quite limited, and it's too precious to give up Bitcoin's robust security and massive liquidity.

Babylon has enabled the use of Bitcoin’s security in two different ways through Bitcoin Timestamping and Bitcoin Staking. The former uses Bitcoin as a timestamp server to prevent transaction reverts or malicious forks, while the latter leverages BTC's powerful liquidity as crypto-economic security, allowing BTC holders to earn additional profits in a native way.

Currently, approximately 55,000 BTC are deposited in Babylon, which is even with a deposit cap set by Babylon. Around 3.9% of the total ETH supply is re-staked on EigenLayer. Considering this, even though BTC holders may be conservative about utilizing BTC, the growth potential of Babylon, with only around 0.2% of the total BTC supply currently staked, is worth considering.

1. Solana Facing Ethereum’s Dilemmas

The Ethereum network, emphasizing decentralization, faces limitations due to its small block size. To improve scalability, Ethereum adopted the rollup-centric Ethereum model, where numerous rollup networks are layered on top of Ethereum L1. However, this approach introduces challenges such as liquidity fragmentation and poor user experience (UX) due to the lack of atomicity across these rollup networks.

In contrast, the Solana network has chosen a more straightforward approach to scalability, relying on hardware advancements to enhance network capacity. While Solana may sacrifice some degree of decentralization compared to Ethereum, its high scalability allows a vast number of transactions to be processed on a single network. This design avoids liquidity fragmentation and ensures strong atomicity, leading to better user experiences.

However, what once seemed like Solana’s unlimited scalability is now meeting its limits. The network recently faced significant congestion, peaking during trading activity around Trump meme tokens. This congestion led to disruptions across Solana's ecosystem, impacting infrastructure and services like Solscan, Phantom, and Jito. Additionally, transaction fees on Solana surged during this period.

Ultimately, Solana is now facing a dilemma similar to Ethereum’s. With increasing network activity driving higher demand for block space, Solana’s network has become congested. This has sparked discussions within the Solana community about various scalability solutions, including Solana-based Layer 2s. Factors limiting Solana’s scalability are numerous, and Solayer aims to address these issues through restaking and a redesigned SVM.

2. Solayer: Improving UX, Liquidity, and Scalability

Solayer is a restaking platform in the Solana ecosystem, backed by prominent venture capital firms such as Binance Labs, Polychain Capital, Hack VC, and Maelstrom. Unlike Ethereum-based restaking platforms like EigenLayer or Symbiotic, Solayer takes a unique approach.

Where traditional restaking platforms allow users to restake native tokens into external protocols, Solayer enables SOL tokens to be restaked into dApps (endoAVS) within the Solana ecosystem. (In the future, Solayer plans to support restaking into external protocols as well.) This is made possible by Solana's existing Stake-weighted Quality of Service (swQoS) mechanism, which allows dApps to process users' transactions more reliably and quickly through restaking. This will be explored in greater detail in Part 3.

Additionally, Solayer issues sUSD, a stablecoin backed by U.S. Treasury Bills, through a decentralized RFQ system. This sUSD can also be utilized as a restaking token for external protocols (exoAVS). Beyond this, Solayer has revealed its vision for Solayer Chain, based on the newly enhanced infiniSVM, to address Solana L1’s scalability limitations.

In summary, Solayer aims to improve the UX, liquidity, and scalability of the Solana ecosystem through restaking, sUSD, and infiniSVM. Let’s dive into each aspect.

3. SOL Restaking for EndoAVSs

3.1 Solana’s swQoS

To understand Solayer's restaking, it is essential first to understand Solana's stake-weighted Quality of Service (swQoS). swQoS is a mechanism designed to effectively handle the high transaction volume on the Solana network.

(Source: Helius)

Solana’s swQoS prioritizes the transmission of transactions to block leaders based on the stake held by validators. This mechanism acts as a type of Sybil resistance mechanism to prevent the network from becoming overloaded by low-staked validators. By prioritizing the transactions of validators with higher stakes, the network is safeguarded against being overwhelmed by malicious, low-quality transactions.

Unlike other networks, Solana does not have a mempool where transactions temporarily reside. When users create transactions, they are sent to RPC nodes and eventually forwarded to the block leader. There are two types of connections to the block leader: a small number of open connections accessible to any RPC node and a larger number of stake-weighted connections accessible only to staked validators. This means that transactions are more likely to be processed if they are forwarded to the leader through staked validators.

From the perspective of RPC nodes, it is advantageous to pair with staked validators to process user transactions more effectively. Validators can virtually allocate a portion of their SOL stake to specific RPC nodes, granting them access to stake-weighted connections. For users, sending transactions to RPC nodes paired with staked validators reduces the likelihood of transaction delays or failures, thereby improving the UX.

Although swQoS and priority fees both aim to enhance UX, they operate at different stages of the transaction lifecycle. Priority fees are a factor when the leader determines transaction priority after receiving them, while swQoS increases the likelihood of transactions reaching the leader in the first place.

3.2 Endogenous AVS

Unlike traditional restaking platforms, Solayer enables restaking for dApps on the Solana network. This is designed to leverage Solana’s swQoS mechanism to allow dApps to process user transactions faster and more reliably. dApps participating in restaking are referred to as Endogenous AVS (endoAVS) to distinguish them from external protocols participating in traditional restaking platforms (Exogenous AVS; exoAVS).

(Source: Solayer)

The restaking process in Solayer follows these steps:

1. Users stake SOL in the Solayer restaking manager and receive sSOL, a corresponding liquid staking token.

2. The SOL received from users is delegated to Solayer-recommended validators.

3. Users delegate their sSOL to endoAVS.

4. endoAVS issues AVS tokens corresponding to the delegated sSOL to users, serving as stake proof to retrieve staked SOL and claim rewards.

(Source: Solayer)

The dApps receiving sSOL delegations from users can process transactions more quickly and reliably thanks to Solana's swQoS mechanism. When executing transactions, dApps send them to Solayer-specified RPC nodes, which forward the transactions to delegated validators. Due to the swQoS mechanism, these transactions have better access to the leader than regular transactions, resulting in improved UX for endoAVS users.

3.3 Super Liquidity

(Source: Solayer)

An interesting aspect of Solayer’s restaking model is that, unlike traditional restaking platforms where staked tokens directly participate in the PoS process of external protocols and require an unbonding period for unstaking, Solayer’s restaking allows immediate unwrapping into sSOL because it is delegated to dApps. Even if numerous endoAVSs exist within the Solayer ecosystem, their corresponding AVS tokens can be immediately unwrapped into sSOL. As a result, only a single sSOL-SOL pool is required, providing a significant liquidity advantage.

4. sUSD

4.1 Minting Process

Solayer provides the sUSD stablecoin, which offers stable yields and can later be utilized for restaking. sUSD is based on U.S. T-bills and enables minting and redemption through an on-chain, non-custodial decentralized RFQ system.

(Source: Solayer)

The Solayer RFQ protocol is a decentralized system that matches users' USDC quotes with qualified tokenizers. All processes are fully automated through smart contracts.

The sUSD minting process is as follows:

1. Users lock USDC in the RFQ protocol and create a quote specifying the amount, expiry time, and commission rate.

2. Qualified Liquidity Providers purchase T-Bills using the user's USDC and send wrapped T-Bills (tokenized T-Bills) to the RFQ protocol.

3. The RFQ protocol transfers the wrapped T-Bills to the sUSD program, where they are locked.

4. The sUSD program issues sUSD corresponding to the value of the locked wrapped T-Bills and delivers it to the user.

5. If desired, users can restake sUSD in exoAVS to earn additional yields.

The redemption process for sUSD is essentially the reverse of the minting process. When users send sUSD to the sUSD program, the program sends an equivalent value of wrapped T-Bills to the RFQ protocol. Qualified Liquidity Providers fulfill the redemption order by transferring USDC to the RFQ protocol, and the user ultimately receives USDC.

4.2 Token 2022 Interest Bearing Extension

How does sUSD accumulate T-Bill yields? sUSD is based on the reference implementation of the Token 2022 interest-bearing extension.

Solana's Token-2022 is an enhanced version of the original SPL Token program, introducing new functionalities through various extensions. One notable feature is the Interest-Bearing Tokens extension, which allows tokens to accrue interest over time.

This extension enables tokens to increase or decrease in value over time, similar to how a bank savings account accumulates interest. The interest is continuously compounded based on the network timestamp, allowing for seamless growth in token value. Importantly, this interest accrual is reflected in the token's displayed amount without actually minting new tokens, preserving a stable token supply.

Due to Solana's account model constraints, directly minting additional tokens to all holders is challenging. Instead, the interest-bearing extension modifies the "multiplier" of the holding amount to reflect interest accumulation. This means that the token amount displayed in a wallet is calculated by multiplying the actual holding amount by a scaling factor that accounts for the accrued interest. As a result, the balance appears to grow over time, much like how a bank account balance increases with interest, without the need to mint new tokens.

5. Solayer Chain

5.1 Transaction Process

Following restaking and sUSD, Solayer recently introduced the Solayer Chain, built on the hardware-accelerated SVM framework known as infiniSVM. The goal is to achieve 1M+ TPS.

(Source: Solayer)

The transaction processing method of Solayer Chain is highly similar to Solana's:

1. Signature verification: Verifying the transaction signature.

2. Deduplication: Filtering out duplicate transactions that have already been processed or contain the same data.

3. Scheduling: Determining the order in which transactions are processed.

4. Banking: Executing transactions and applying results to the blockchain state.

5. Storage: Storing the results of processed transactions.

5.2 infiniSVM vs. SVM

5.2.1 Microservice Architecture

Solayer Chain's infiniSVM introduces several improvements to Solana's SVM. The first is a microservice architecture. While Solana processes the above steps sequentially in a single execution environment, infiniSVM separates the signature verification, deduplication, and storage steps into independent processes that run on multiple servers simultaneously. This increases overall system scalability.

5.2.2 Simulation Stage

The second improvement is the addition of a simulation stage in the transaction pipeline. This stage can also be performed in parallel across multiple nodes. During this stage, pre-execution simulations are conducted to minimize transaction conflicts. Transactions are executed against the most recently committed state, and intermediate execution snapshots are captured. Additionally, the Winter-Holt DESP predictive model is used to further reduce transaction conflicts. If a conflict is detected, the system can recover the state using the nearest valid snapshot stored during the simulation, avoiding the need to re-execute the entire transaction and reducing processing time.

5.2.3 Transaction Semantic Aware Scheduling

The third improvement is more detailed analysis of read/write sequences before transaction execution compared to SVM. On the Solana network, transactions accessing different accounts are processed in parallel. However, Solana only considers which accounts are accessed by a transaction without distinguishing between read and write operations. For example, transactions that only read from account A and those that only write to account A could theoretically be processed in parallel, but Solana treats them as potential conflicts. InfiniSVM optimizes this by maximizing parallel execution possibilities.

5.2.4 Remote Direct Memory Access (RDMA)

The fourth improvement is the use of RDMA for data transfer between Solayer Chain nodes. When computers exchange data, it usually involves the operating system, which increases processing time and CPU load. RDMA eliminates this by directly transferring data from the memory of one computer to another's memory. This is made possible by using special network devices called RNICs instead of standard network cards. By bypassing the operating system, RDMA significantly reduces CPU workload and accelerates data transfer.

6. Final Thoughts

We have explored Solayer's three core features: endoAVS restaking, the sUSD stablecoin, and infiniSVM. At first glance, these features may seem unrelated. For instance, there is no publicly disclosed connection between infiniSVM and restaking yet. However, these features collectively address the UX, liquidity, and scalability challenges within the Solana ecosystem, empowering users to interact with the Solana ecosystem more effectively.

“If you survive until tomorrow, it could mean that either a) you are more likely to be immortal or b) that you are closer to death.” — The Black Swan by Nassim Nicholas Taleb

In November 2022, FTX filed for bankruptcy, and two years have already passed since then. The market, as if forgetting the collapses of FTX and Terra, has recovered rapidly, with Bitcoin now approaching the $100,000 mark.

While the market's steady progress is exciting, as an investor who entered the crypto market in 2020, I’ve witnessed so many crash events that I can’t help but worry about what the next black swan event might be.

The collapse of FTX and Terra had different causes, processes, and outcomes, but they share one clear similarity: the cycle of virtuous and vicious loops. Both were structured in a way that allowed for explosive growth in favorable market conditions but also endless downward spirals when things turned sour.

Although not as significant as FTX or Terra, a similar characteristic can be currently observed in another entity: MicroStrategy. Personally, I believe MicroStrategy carries less risk compared to the other two, but its practice of leveraging debt to purchase Bitcoin is not without danger. I'd like to briefly discuss this.

1. Status Quo

1.1 Company Overview

MicroStrategy was originally established as a business intelligence company by Michael Saylor in 1989. In its early years, the company focused on analytics software, eventually expanding into mobile applications and cloud-based services in line with technological advancements.

The company went public in 1998 with a valuation of approximately $1 billion. Aside from a brief surge during the dot-com bubble, MicroStrategy's market value hovered around $1 billion, making it a relatively unremarkable stock up until 2020. However, everything changed on August 11, 2020, when MicroStrategy, as a publicly traded company, announced its first Bitcoin purchase. Since then, it has consistently added BTC to its holdings, driving its market capitalization to an exponential growth of around $90 billion today.

1.2 Where Do Funds Come From?

MicroStrategy has acquired a total of 402,100 BTC at an average price of $58,263, representing approximately 1.9% of Bitcoin's total supply of 21 million—a staggering amount. The company primarily funds its Bitcoin purchases through three main methods:

1.2.1 Cash Reserves

As a software company, MicroStrategy generates revenue through product licenses, subscription services, and product support. According to its 10-K reports, the company generated ~$499M in revenue in 2022 and ~$496M in 2023 from its core business.

After deducting the cost of revenues and operating expenses (excluding cryptocurrency price fluctuations), MicroStrategy was left with $10.5M in 2022 and $0.8M in 2023. These funds can be used for purchasing BTC or paying interest on existing debt.

1.2.2 Convertible Senior Notes & Stock Offering

MicroStrategy’s primary method of raising funds is by offering convertible senior notes. As of Q3 2024 Financial Results, the company has accumulated ~$4.26B in debt through such notes, with an average annual interest rate of 0.811%, resulting in $34.6M in yearly interest payments.

Recently, on November 21, 2024, MicroStrategy offered an additional $3B in convertible senior notes. This brings its total debt to over $7B. Notably, the new $3B offering carries a 0% coupon, meaning no interest expense is incurred. Instead, investors can convert these notes into stock at a 55% premium in the future.

Besides convertible notes, MicroStrategy also conducts stock offerings. In Q3 2024, the company issued $1.1B in Class A common stock.

Looking ahead, MicroStrategy announced plans to raise an impressive total of $42B over the next three years from 2025 to 2027. Of this, $21B will come from equity offerings, while the remaining $21B will be raised through fixed-income offerings. The timeline includes raising $5B annually in 2025, $7B in 2026, and $9B in 2027 from each funding source.

1.3 Statistical Highlights

Here are some interesting data and facts about MicroStrategy:

• The average purchase price of Bitcoin acquired by MicroStrategy is $58,263. The total market value of its holdings is $38.5B, while the carrying value is $23.4B, resulting in approximately $15B in unrealized gains.

• MicroStrategy’s market valuation is trading at roughly 2.2 times the value of its Bitcoin holdings. This valuation disparity has been a target for short-selling by funds like Citron Research.

• Despite the increasing number of shares due to equity offerings, the BTC per basic share has actually risen. This is because the rate at which MicroStrategy is acquiring Bitcoin surpasses the dilutive effect of additional share issuances.

2. Is Microstrategy The Next FTX?

While structural issues played a role in the collapses of FTX and Terra, the moral hazard of their founders also significantly contributed. By definition, black swan events are inherently unpredictable, making it impossible to assess whether there is any moral hazard within MicroStrategy based on publicly available information. Therefore, the focus here will be on analyzing structural risks.

Although the title of this discussion may sound grand, the structural risk associated with MicroStrategy is actually straightforward: Bitcoin investment through leverage. If the company were only investing its own equity in Bitcoin, the impact of a price crash would be relatively minimal.

However, as highlighted in the Q3 2024 Report, MicroStrategy’s goal is to use prudent leverage to acquire as much BTC as possible, enhance shareholder value, and outperform Bitcoin itself.

As everyone knows, leverage is a double-edged sword. It inherently comes with interest costs on debt, and if the value of BTC purchased with leverage declines, the company may be forced to sell its holdings to repay creditors.

Michael Saylor has repeatedly stated in the media that he has no intention of selling BTC, which means any forced liquidation could severely impact the company’s value. This would likely lead to a decline in MSTR’s stock price, difficulty in future funding, and the potential collapse of its leverage strategy, with significant ripple effects on the market.

To evaluate whether MicroStrategy’s leverage strategy is sustainable, I’m gonna examine two key aspects:

• Interest Costs

• Sustainable Volatility

2.1 Can MicroStrategy Manage Its Interest Costs?

First, let’s consider the interest costs. With the recent issuance of $3B in convertible senior notes at a 0% coupon, MicroStrategy’s total debt stands at approximately $7.3B, with an average interest rate of 0.476%, resulting in $34.6M in annual interest payments. The key question is: can MicroStrategy continue to manage these payments?

According to the company’s 2023 10-K filing, excluding digital assets, MicroStrategy holds approximately $1.13B in other assets. This is significantly higher than the annual interest payment of $34.6M, suggesting that the company should have no immediate issues covering interest costs.

Additionally, as noted earlier, MicroStrategy generates substantial revenue from its core business of business intelligence software. However, a concerning factor is that after deducting the cost of revenues and operating expenses, the remaining profit is relatively small and has been declining over time.

The scale of future debt also cannot be overlooked. As mentioned, MicroStrategy plans to issue an additional $21B in convertible senior notes over the next three years. This would increase its total debt to $28.3B. Applying the current average interest rate of 0.476%, annual interest payments could rise to $134.7M, a potentially unsustainable amount in the long term.

While MicroStrategy has reduced its interest burden by issuing 0% coupon notes recently, it remains uncertain whether the company can continue to secure such low-interest debt. This issue will be explored further in the next section.

2.2 Is Future Fundraising Sustainable?

2.2.1 Will Investors Continue to Provide Funding?

The first question is the sustainability of fundraising. While the current positive sentiment in the cryptocurrency market makes it seem easy to secure funding, historical data shows that between February 2021 and March 2024, MicroStrategy did not issue any convertible senior notes. This corresponds to the period from Bitcoin’s first crash in 2021 to the recent market recovery.

While there’s no definitive evidence, Bitcoin’s price cycle has historically aligned with its 4-year halving cycle. This suggests a high likelihood of a bear market around 2026–2027, with the BTC long-term power law model indicating a potential price floor of $53K–$70K during that time. Considering MicroStrategy’s average Bitcoin acquisition price of $58K, a bear market could make securing additional funding more challenging.

2.2.2 Volatility Matters

Beyond Bitcoin’s price, volatility plays a critical role in the sustainability of funding through convertible senior notes.

Some readers may wonder: Why does MSTR stock trade at a premium over its net asset value (NAV) by more than 2x? Why would investors participate in the recent $3B convertible senior notes offering with a 0% coupon rate and a 55% premium?

The key to these questions lies in volatility. Bitcoin and MSTR exhibit significantly higher volatility compared to other assets, making them attractive to investors. MicroStrategy even highlights its stock’s volatility in its Q3 2024 IR report.

High volatility enables various trading strategies, such as delta hedging, gamma trading, and volatility arbitrage.

• Delta measures the sensitivity of an option’s price to changes in the underlying asset's price.

• Gamma measures how much delta changes with movements in the underlying asset’s price.

• Gamma trading leverages these changes to profit from market volatility, making higher volatility more lucrative.

As Bitcoin is inherently more volatile than traditional stocks, and MSTR amplifies this volatility through leveraged Bitcoin purchases, MSTR stock and its convertible senior notes attract significant interest from hedge funds. These notes act not only as a debt instrument but also as a call option, allowing the holder to convert them into shares at a specific price, further increasing their appeal.

For MicroStrategy to maintain smooth fundraising, the volatility of BTC and MSTR must persist. However, with the approval of Bitcoin ETFs and the increasing presence of institutional investors, the market is becoming more stable, leading to a gradual decline in volatility. If Bitcoin’s volatility decreases, MSTR’s NAV premium may shrink, affecting its stock price and reducing the attractiveness of its convertible senior notes for fundraising.

3. In the End, Price Matters

Ultimately, the most critical key factor is the price of Bitcoin. Had the cryptocurrency market not experienced a downturn, FTX could have become a massive exchange rivaling Binance. Similarly, if Curve Finance's UST pool had not been attacked, Terra might have emerged as the third-largest mainnet after Bitcoin and Ethereum.

If BTC's price continues to rise steadily, MicroStrategy's current strategy could create an endless virtuous cycle, driving explosive growth for both the company and the cryptocurrency market. However, if BTC’s price crashes, the worst-case scenario could see MicroStrategy selling its Bitcoin holdings to repay debt, initiating a vicious cycle.

The silver lining is that MicroStrategy's debt is unsecured. The company had previously issued bonds secured by Bitcoin but fully repaid them by Q3 2024. If Bitcoin were still collateralized, a forced liquidation in the event of repayment difficulties could have been disastrous.

In my personal opinion, MicroStrategy does not appear to face significant immediate risks. Its interest costs remain manageable, and fundraising has been smooth thus far. However, the scale of planned future fundraising is unprecedented, and the volatility of Bitcoin, the underlying asset, is expected to decrease. Therefore, I anticipate that the level of risk will be much higher between 2025 and 2027.

Bitcoin’s value is rapidly increasing, positioning itself as a challenger to gold’s status. The question remains: will MicroStrategy continue acquiring BTC in a sustainable manner to become one of the world’s strongest companies? Or will it serve as another cautionary tale, trapped in the narrative of a tulip bubble? Only time will tell.

1. Data Privacy Matters

1.1 The Double-Edged Sword of AI

Since the launch of ChatGPT, powered by GPT-3.5, on November 30, 2022, the world has been abuzz with talk of AI. AI services have become so deeply embedded in daily life that they’re now almost indispensable. Major tech companies like Google and Microsoft are expected to exceed $250 billion in AI-related spending by 2025. Even as I write this, I’m relying on AI tools for research and translation.

While AI undeniably promises massive productivity gains and immense wealth for humanity, it’s equally critical that we manage this unprecedented technology wisely and minimize its potential downsides. If humanity fails to control AI—and, further down the line, AGI—society could collapse entirely, as warned by Leopold Aschenbrenner in “Situational Awareness.”

1.2 AI Safety

Moreover, if we move in the direction of making machines which learn and whose behavior is modified by experience, we must face the fact that every degree of independence we give the machine is a degree of possible defiance of our wishes. - Nobert Wiener (1949)

AI safety focuses on ensuring stability and mitigating potential risks throughout the development, deployment, and application of AI technologies. While the benefits of AI are vast, numerous experts have voiced concerns about its dangers—an ongoing topic of debate for years.

From Geoffrey Hinton, the “Godfather of AI” and 2024 Nobel Physics laureate, to figures like Elon Musk, Bill Gates, and Stephen Hawking, leaders across various fields have warned about the risks of AI. Even within OpenAI, tensions between rapid AI advancements and robust safety measures led to the resignations of Ilya Sutskever and Mira Murati this year.

As Delphi Digital highlighted in “DeAI I: The Tower & the Square,” AI tends to reinforce traditional hierarchies (the "tower"). For this reason, achieving AI safety requires interdisciplinary efforts spanning social systems, technology, and policy to prevent misuse or malicious applications. So, what does AI safety entail?

• Robustness & Reliability: Ensuring AI systems operate consistently and do not fail when encountering unexpected inputs.

• Fairness & Bias Mitigation: Identifying and eliminating biases during training to produce equitable outcomes.

• Transparency & Explainability: Enhancing clarity about how AI generates outputs, making processes understandable to users.

• Accountability & Governance: Establishing clear responsibilities for AI’s development and use, and building governance systems to address unforeseen issues.

• Privacy & Security: Protecting user data and safeguarding AI systems from malicious attacks.

• Alignment with Human Values: Designing AI to act in alignment with human ethics and values, minimizing risks of harm.

1.3 Why Data Privacy Matters

Among the various aspects of AI safety, this article focuses on data privacy. While most people understand the importance of protecting their data, many remain unaware of the risks associated with its misuse. Instead of safeguarding their personal information, users often place blind trust in big tech platforms, exposing vast amounts of private data online. This leaves them vulnerable to cyberattacks, with data breaches escalating rapidly year by year.

In the age of AI, data is indispensable. Developing more advanced and specialized AI models requires an immense volume of high-quality data. As AI evolves, the importance of data privacy will only grow, for several key reasons:

• Cases of unauthorized data usage are already emerging. For example, OpenAI and Microsoft have faced lawsuits for allegedly exploiting vast amounts of user data without consent. Similarly, Meta has faced comparable accusations.

• Many existing AI models have been trained on nearly all accessible online data. To develop even more powerful models, the demand for additional data—whether synthetic or personal—is only increasing.

• We are now entering the era of AI agents, where personalized AI agents tailored to individuals will boost productivity. However, these agents require training on users' personal data. Unless this training happens offline in local environments, it will inevitably raise serious concerns about data privacy.

The fundamental challenge for the AI industry is clear: How can we train models on personal data without infringing on privacy? While laws and regulations aim to mitigate these risks, the numerous reported breaches indicate that prevention at the technological level is the most effective solution.

2. Enter Nillion

2.1 Overview

Nillion Network provides a solution that leverages various Privacy-Enhancing Technologies (PETs) to securely encrypt and store user data, while also enabling computations on the encrypted data. For instance, in the context of AI training, Nillion allows sensitive and confidential personal information to be used safely for training AI models without ever exposing the data to external parties.

Traditionally, processing high-value data such as personal information involved decrypting the encrypted data, performing computations, and then re-encrypting it. This approach was not only inefficient but also introduced security risks during the decryption process. Nillion overcomes these limitations by utilizing PETs like Multi-Party Computing (MPC), enabling computations on encrypted data without needing to decrypt it, thereby maintaining security and efficiency.

2.2 What Are PETs?

Privacy-Enhancing Technologies are tools designed to protect data while allowing secure computation. Some of the most prominent PETs in the blockchain ecosystem include Zero-Knowledge Proofs (ZK), Fully Homomorphic Encryption (FHE), Trusted Execution Environments (TEE), and Multi-Party Computation (MPC). While Nillion initially focuses on MPC, it plans to integrate additional PETs in the future. Let’s explore the unique features and differences among these technologies.

2.2.1 ZK

Zero-Knowledge technology focuses on proving the validity of data without revealing the data itself. A data owner can decrypt their data locally, make changes, re-encrypt it, and then generate a ZKP to demonstrate the validity of these changes to another party without exposing the actual data. In the blockchain ecosystem, Zero-Knowledge technology is applied in various ways:

• Private Transactions: Users can conduct transactions involving their assets privately, processing them locally and submitting a ZKP to the network. This allows for transactions without exposing wallet balances or transaction history (e.g., Zcash, Iron Fish).

• Scalability Solutions: ZK technology enhances blockchain scalability by generating single, succinct ZKPs for complex off-chain computations. These proofs can be verified on-chain, enabling efficient execution of operations that would otherwise be impractical on-chain (e.g., ZK rollups, ZK coprocessors).

• Identity Verification: Sensitive personal data, such as identity information, can be validated without being disclosed to other parties (e.g., Polygon ID).

2.2.2 FHE

Fully Homomorphic Encryption (FHE) enables data owners to delegate computations on their data to a third party without revealing the actual data. In this process, the third party can perform operations directly on the encrypted data without needing to decrypt it. The resulting encrypted output can then be decrypted by the data owner using their private key. While FHE is a promising technology, its maturity level in the blockchain ecosystem remains relatively low. Notable projects exploring FHE include:

• Fhenix: An Ethereum Layer 2 solution based on FHE technology. It allows developers to create encrypted smart contracts using Solidity and execute encrypted computations on data.

• Zama: Zama’s fhEVM coprocessor enables developers to easily build confidential smart contracts on Ethereum.

2.2.3 TEE

Trusted Execution Environment (TEE) refers to a secure hardware-based area designed to process sensitive data safely without exposing it externally. Through an attestation process, it can verify that specific code is running securely within the TEE. Blockchain projects leveraging TEE technology include:

• TEE-Boost: Flashbots’ TEE-Boost removes centralized relays in MEV-Boost by ensuring builders use verified software.

• TEE Proof for Rollups: Some zk rollups, including Taiko, utilize TEE proofs as part of their multi-proof systems. TEE proofs verify the validity of computations through attestation, making them suitable for rollup proofs.

• Privacy: Secret Network uses Intel SGX CPUs in its validators to process user transactions privately, ensuring secure handling of sensitive data.

2.2.4 MPC

Multi-Party Computation (MPC) is a technology that enables computations on inputs from multiple parties without revealing the original data. A classic example is Andrew Yao's “The Millionaire Problem,” where two millionaires use MPC to determine who is wealthier without disclosing their actual wealth.

In the blockchain ecosystem, one of the most prominent applications of MPC is the MPC wallet. These wallets split a private key into multiple shards, which are managed by different parties. No single party can reconstruct the full key, ensuring security. Even if one shard is compromised or lost, the remaining shards can maintain functionality and allow wallet recovery. Custodial services like Fireblocks use MPC technology to securely store crypto assets.

2.3 Nillion’s Novel MPC Protocol

Nillion Network leverages MPC to enable blind computing on private data. Its MPC protocol offers two major advancements compared to traditional MPC technologies:

• Support for Non-linear Arithmetic

• Asynchronous Computation

2.3.1 Support for Non-linear Arithmetic (Sum of Products)

Most MPC protocols are designed for linear operations like addition (e.g., x+y) or multiplication by a constant (e.g., x*3). Nillion’s MPC protocol, however, extends this capability to support non-linear operations, such as the sum of products.

The "sum of products" refers to operations where multiple terms are multiplied and then added together. For example, ab+cd. Here, a, b, c, and d are called factors, while their products ab & cd are referred to as terms.

2.3.2 Asynchronous Computation Through Pre-processing

Traditional MPC protocols often require interactive communication between parties during computations, making them dependent on synchronous environments. Nillion’s MPC protocol eliminates this limitation by introducing a pre-processing phase.

During pre-processing, the network nodes generate masks for each factor and term in the sum of products operation. These masks are shared with users in advance, enabling them to create hidden inputs that are later used for blind computation. This approach eliminates the need for real-time communication during computation, enabling asynchronous execution while improving efficiency and speed.

This innovative approach significantly enhances the performance and usability of MPC protocols, paving the way for broader applications. Detailed workflows and code examples will be explored in the next section.

2.4 Nillion’s MPC Protocol Workflow

The workflow of Nillion’s MPC protocol can be divided into two main phases:

1. The pre-processing phase, where masks for the factors and terms in a sum-of-products expression are generated.

2. The non-interactive computation phase, where computation is performed without real-time communication between nodes.

2.4.1 Assumptions

For this example, we assume the following setup:

• There are 3 contributors (a, b, c) providing private inputs.

• There are 3 nodes (0, 1, 2) performing blind computation.

• The sum-of-products expression to compute is (1x2x3)+(4x5).

Let’s walk through how Nillion’s MPC protocol handles this computation. For illustration, Python code is shown to demonstrate how a non-interactive MPC protocol works.

nodes = [node(), node(), node()]

2.4.2 Setting the Workflow Signature

Contributors must first agree on the workflow signature, which specifies the number of factors in each term of the sum-of-products expression. For the expression (1x2x3)+(4x5), the signature is [3, 2]. This signature is shared with all nodes.

signature = [3, 2]

2.4.3 Pre-processing Phase

Using the provided signature, nodes perform pre-processing to generate random masks for contributors’ inputs. These masks ensure the privacy of the inputs during computation.

preprocess(signature, nodes)

2.4.4 Assigning Inputs

Contributors assign their inputs to specific positions in the sum-of-products expression. Positions are represented as (term_index, factor_index).

coords_to_values_a = {(0, 0): 1, (1, 0): 4}
coords_to_values_b = {(0, 1): 2, (1, 1): 5}
coords_to_values_c = {(0, 2): 3}

2.4.5 Requesting and Applying Masks

Each contributor requests masks for their assigned positions from the nodes. Using the node.masks() function, contributors receive the masks from each node and apply them to their input values using the masked_factors() function. Below is an example for contributor a; contributors b and c follow the same process.

masks_from_nodes_a = [node.masks(coords_to_values_a.keys()) for node in nodes]
masked_factors_a = masked_factors(coords_to_values_a, masks_from_nodes_a)

2.4.6 Broadcasting Masked Values

Contributors broadcast their masked inputs to all nodes.

broadcast = [masked_factors_a, masked_factors_b, masked_factors_c]

2.4.7 Nodes Perform Blind Computation

Each node uses the masked values to compute its assigned share of the result locally.

result_share_at_node_0 = nodes[0].compute(signature, broadcast)
result_share_at_node_1 = nodes[1].compute(signature, broadcast)
result_share_at_node_2 = nodes[2].compute(signature, broadcast)

2.4.8 Calculating the Final Result

Nodes aggregate their computed shares to produce the final result.

final_result = int(sum([result_share_at_node_0, result_share_at_node_1, result_share_at_node_2]))

2.4.9 Summary

Although the process may seem complex, it’s actually quite straightforward. Contributors define the structure of the computation, and nodes generate random masks during the pre-processing phase. Contributors then receive these masks from the nodes, apply them to their inputs to ensure privacy, and send the masked inputs back. The nodes use the masked inputs to perform the computation and produce the final result.

Because the inputs are masked throughout the process, contributors’ data remains private. Additionally, the pre-generated masks allow for asynchronous execution, making the computation more efficient and scalable.

2.5 Nillion Network Architecture

Now that we understand how Nillion leverages MPC and other PETs to enable blind computation, let's delve into the network's architecture. Nillion adopts a dual network architecture to achieve its goals:

2.5.1 Orchestration Layer (Petnet)

Protocols like MPC, ZK, FHE, and TEE each have specific strengths and limitations when it comes to privacy, and they complement each other in addressing different needs. The Orchestration Layer is designed to securely process data by enabling nodes to utilize Nillion's suite of PETs. It operates purely as a protocol layer and is not a blockchain network.

2.5.2 Coordination Layer (NilChain)

The Coordination Layer is a blockchain network built on the Cosmos SDK, facilitating tasks such as payment processing and resource coordination for the Orchestration Layer. Thanks to its Cosmos SDK foundation, the Coordination Layer enables seamless interaction with other networks via IBC.

The native token, NIL, plays a key role in this layer, serving as:

• The governance token for decision-making within the network.

• The reward mechanism for nodes participating in Petnet operations.

• A fee token for gas costs and blind computation requests by users on the Nillion Network.

2.6 Developer Tools

To simplify onboarding for developers, Nillion offers a comprehensive set of tools through its Nillion SDK, specifically designed for dApp developers. The SDK includes:

• Nillion CLI Tool: A command-line interface enabling interaction with the network, such as checking its status, uploading programs, and storing confidential data.

• Local Network: An environment for testing the network locally.

• User and Node Key Generation: Tools to generate keys for users and nodes in local environments.

• Program Simulator: A feature that allows developers to simulate Nada programs locally before deploying them to the network.

• Nada Compiler: A compiler for programs written in Nada.

Nada, a domain-specific language (DSL) used within the Nillion Network, is designed specifically for privacy-focused applications. Its syntax is inspired by Python, making it accessible and familiar to developers.

Nillion also provides open-source libraries such as Nada Numpy and Nada AI to streamline the creation of privacy-preserving AI systems. These libraries adapt popular Python tools to work seamlessly with the Nada DSL.

• Nada Numpy: Enables the use of most NumPy functions in the Nada environment.

• Nada AI: An all-in-one AI solution that facilitates integrating models from other AI frameworks into the Nillion ecosystem. It supports uploading programs, storing data, and running privacy-preserving inferences within the network.

With these tools and libraries, Nillion empowers developers to build innovative, privacy-centric applications with ease.

3. Nillion Ecosystem

Nillion Network offers a unique concept of blind computation powered by various PETs. Its modular nature allows other projects to integrate and utilize the network, making Nillion a versatile tool for a wide range of applications.

3.1 Blockchain Networks

Most blockchain networks do not inherently provide private storage or blind computing capabilities. If these functionalities were integrated into mature ecosystems, it could enable groundbreaking applications such as private AI assistants, confidential DeFi, and secure gaming.

Developers and dApps from the following blockchain ecosystems can delegate tasks like private data storage or blind computation to Nillion. This effectively makes Nillion a modular blind computing layer for these networks. Notably, NEAR Protocol, with its focus on AI applications, could benefit significantly in areas like private inference, synthetic data, and federated learning.

• Aptos

• NEAR Protocol

• Arbitrum

3.2 DeFi Protocols

• Kayra: A decentralized dark pool exchange utilizing MPC technology for its order book.

• Choose K: Offers encrypted order books powered by Nillion’s MPC technology.

3.3 Infrastructure

• zkPass: An oracle protocol combining ZKP, TLS, and MPC to enable on-chain verification of private internet data.

• Skillful AI: Provides customized virtual assistants by leveraging Nillion’s blind computing layer to securely handle high-value data.

• Coasys: Focuses on eliminating data silos and strengthening data sovereignty using Nillion’s secure data processing capabilities.

3.4 AI Infrastructure

• Ritual: A decentralized AI infrastructure using Nillion’s blind computation for private model inference and secure storage.

• Rainfall: Plans to launch a personalized AI platform and data marketplace in collaboration with Nillion.

• Mizu: A GitHub-like service for datasets that uses Nillion’s blind computation to enable private data repositories.

• Nesa: A lightweight L1 network for AI inference, collaborating with Nillion on privacy and inference-related tasks.

• Nuklai: An L1 network specializing in AI-related data.

3.5 Consumer AI Services

• Aloha: An AI-powered dating app that uses Nillion to protect user data while enabling secure matchmaking.

3.6 AI Agents

• Virtuals Protocol: A platform for creating AI agents for gaming and entertainment, leveraging Nillion’s secure infrastructure for private training and inference.

• Crush: A crypto hub using autonomous AI agents to simplify interactions with blockchain ecosystems. It employs Nillion for secure learning and personalized suggestions.

• Capx: A user-centric AI infrastructure allowing developers to build, monetize, and trade AI agents securely with Nillion.

• Verida: An open-source layer providing private data and confidential computing for secure AI assistants.

3.7 Custody

• Salt: A platform leveraging Nillion to let users maintain ownership of their assets while enabling management through AI or other asset managers under pre-set policies.

3.8 Secure Storage

• NilQuantum: Offers quantum-secure password storage for Nillion ecosystem users.

• Flux: Enables private and secure document storage.

• Dwinity: Manages user accounts and data securely in collaboration with Nillion.

• Blerify: A project focused on verifiable credentials.

3.9 Healthcare

• Maya: Collaborates with Nillion to build storage and secure computation infrastructure for healthcare data.

• Monadic DNA: Uses blind computing technology to securely process personal DNA data.

• Space of Mind: An online mental health counseling platform.

• Agerate: Offers diagnostic services based on blood samples, using Nillion for secure patient data management.

3.10 Others

• DecentDAO: Provides tools for DAOs, including private voting and secure data management powered by Nillion.

• Mailchain: Facilitates private communication using Nillion.

4. Final Thoughts

We are at the forefront of an inevitable AI revolution. While AI, led by big tech, promises to enhance our lives, it also raises significant data privacy concerns. Many companies have already adopted encryption technologies to address these challenges. However, Nillion goes a step further by incorporating blockchain to create a truly decentralized blind computing layer with incentivization mechanisms.

One concern I have about Nillion is whether the community truly recognizes the importance of data privacy at this point. While blind computing is highly likely to become an essential stack in the AI industry in the distant future, the current AI ecosystem is still in its infancy. Many users seem to lack significant awareness of issues related to personal data. It will be interesting to see if Nillion’s technological vision aligns with the perspectives of society and regulators on AI safety. If these align, it could pave the way for a future where our society moves closer to an AI-driven utopia.

1. Everything, Everywhere, All at Once

Oracles have become an essential infrastructure in the blockchain ecosystem. They act as middleware, enabling secure utilization of various off-chain data in on-chain environments. Oracles integrate all types of data sources (Everything) and provide data across all on-chain ecosystems (Everywhere), including mainnets, DeFi, and prediction markets simultaneously (All at Once).

While it is often claimed that blockchain oracles handle all types of data, in reality, they primarily deal with data applicable to DeFi. Oracles supply accurate price data from sources like market makers, centralized exchanges, and websites, and they detect anomalies in smart contracts that constitute DeFi protocols, ensuring their safe and precise operation.

According to data from DefiLlama, the total value secured (TVS) by oracles currently stands at $66.5 billion, which accounts for 60.8% of TVL in the on-chain ecosystem. This figure is direct evidence of the pivotal role oracles play in the blockchain ecosystem.

However, oracles contribute beyond the DeFi ecosystem. In addition to market data, they handle various types of data and play a role in random number generation, such as through VRF (Verifiable Random Function). A notable non-market data use case for oracles is Polymarket, where real-world data, like election results, is fed into smart contracts to determine betting outcomes and facilitate fund transfers.

2. Are Oracles Truly Decentralized?

Unlike in the past, oracles have become such fundamental and ubiquitous infrastructure that they’re now considered a mundane narrative in the blockchain ecosystem. But here’s a question: are the oracles we use, knowingly or unknowingly, truly secure and sufficiently decentralized?

Although the narrative around decentralization and security has lost some of its prominence, these qualities remain the greatest advantages of blockchain compared to traditional systems. The ongoing discussions in the Ethereum community about whether Layer 2 networks truly rely on Ethereum’s security and where they stand in terms of decentralization highlight the importance of this topic.

2.1 Why Is the Decentralization of Oracles Less Discussed?

Oracles, which secure over 60% of the total TVL in the blockchain ecosystem, could arguably be considered more critical infrastructure than rollup protocols. So why is there less discussion about decentralization when it comes to oracles? There are some fundamental and psychological reasons:

1. Indirect involvement of funds: Unlike rollup protocols where users directly deposit their funds, oracles don’t directly handle user funds. Instead, they secure a portion of the funds deposited in DeFi protocols. Psychologically, this makes oracle decentralization seem less of a priority compared to rollup decentralization.

2. Nature of data: This is the most fundamental reason. Unlike the source code of rollup protocols, real-world data is inherently centralized. Since retrieving data often relies on centralized sources, decentralization tends to take a backseat.

Despite this, there have been numerous efforts to increase the decentralization of oracle protocols. Greater decentralization in oracles offers the following benefits:

• High uptime: Ensures a continuous supply of data to users without downtime.

• Censorship resistance: Reduces the likelihood of data censorship by providers.

• Reduced attack surface: With more diverse data providers, defending against malicious data manipulation attacks becomes easier. This is especially significant given past cases of centralized oracle attacks exploiting DeFi protocols.

2.2 Evaluation Criteria

To assess the decentralization of oracles currently in active use, I’ve established the following criteria. These are not official standards but rather my framework for reference:

1. How to become a data source or operator: The more permissionless the participation (e.g., staking-based entry), the higher the decentralization.

2. Total Value Secured (TVS) and cryptoeconomic security: A larger cryptoeconomic security base translates to stronger security and greater justification for TVS. If TVS is disproportionately high relative to cryptoeconomic security, it could become a vector for fund-theft attacks.

3. Dispute process and slashing mechanisms: The presence of robust systems to handle disputes or penalize malicious data providers enhances both decentralization and security. A well-implemented dispute process and slashing mechanism are critical for securing oracles.

3. Projects

3.1 Band

• How to become a data source & operator: Validators on Bandchain also serve as oracle operators. To become a validator, BAND tokens must be staked, and only the top 100 validator candidates, based on self-staking and delegated staking, can become active validators.

• Total value secured/cryptoeconomic security: The network secures a total value of approximately $138M from 21 protocols. The value of BAND tokens staked in the network is around $123M. According to Tendermint BFT, the network is at risk if more than one-third of this amount (~$41M) is compromised.

• Dispute process & slashing: Although slashing exists, it applies to validators for downtime, double-signing, and failing to respond to data requests as oracle operators. Validators use a program called YODA to automatically fetch data from data sources when a request is made. However, there is no specific slashing penalty for malicious data submission. Even if a validator acts maliciously, it does not significantly impact the average values submitted by the majority of validators. Still, malicious behavior risks losing reputation in the community and having delegations withdrawn. However, relying solely on reputation to disincentivize malicious data provision is a notable shortcoming.

Band is one of the early oracle projects, based on Bandchain, a blockchain built with the Cosmos SDK. Band aims to provide various data sources beyond price feeds. Band offers four main services:

• Band Standard Dataset: General price feeds fetching cryptocurrency prices from sources like CoinGecko, Binance, Coinbase Pro, and OKX. For FX and commodities, it retrieves data from Fixer, Open Exchange Rates, XE, and AlphaVantage.

• VRF: Random number generation for blockchain services.

• Pricefeed Module: Cosmos SDK chains can integrate Band's Pricefeed Module to send data requests to Bandchain via IBC.

• cw-band: Software enabling Cosmos SDK networks using CosmWasm to utilize Band's data.

3.2 UMA

• How to become a data source & operator: To provide specific data (asserter), you simply need to deposit a bond. Anyone can stake UMA tokens to vote in the dispute process.

• Total value secured/cryptoeconomic security: The network secures a total value of ~$317M from 8 protocols (source: DefiLlama). According to the official website, it has ~$1.43B in TVS. On average, 15M $UMA has participated in voting since 2021, meaning approximately ~$36M would be required to maliciously overturn results.

• Dispute process & slashing: Since it deals with qualitative data, the dispute mechanism is very well-designed. When a dispute arises, voters staking UMA tokens make the judgment, and depending on the outcome, the asserter or disputer may lose or gain their bond.

Unlike other oracles, UMA is more focused on providing data for general statements rather than price feeds. For this reason, it is less commonly used in DeFi protocols and more in contexts requiring qualitative data, such as Polymarket.

UMA employs an Optimistic Oracle model. Initially, an asserter (data provider) stakes a bond and provides data. If a disputer believes the statement is incorrect and submits a dispute with their own bond, the issue is escalated to UMA's Data Verification Mechanism (DVM). In the DVM, voters staking UMA tokens decide the outcome. If the asserter is correct, they receive half of the disputer's bond, and vice versa.

3.3 Redstone

• How to become a data source & operator: Currently, RedStone requires a permissioned approach, involving communication with the RedStone team and meeting legal and technical standards. However, it is expected to gradually shift towards permissionless participation, requiring staking RedStone tokens in the future.

• Total value secured/cryptoeconomic security: The network secures ~$1.7B in total value across 51 protocols. While there is no token yet, RedStone plans to leverage EigenLayer’s AVS in the future for broader cryptoeconomic security.

• Dispute process & slashing: Although RedStone tokens are not yet launched, decentralized juries staking RedStone tokens will eventually participate in disputes. If a data provider is found to have harmed users with incorrect data, users can claim a refund.

RedStone is a rapidly growing oracle gaining popularity among DeFi protocols, especially excelling in LSTs and LRTs. Its key differentiator is providing data on-demand. Unlike traditional oracles, which publish data regardless of usage, RedStone uploads data on-chain based on user demand, making it cost-efficient.

Currently, it offers price feeds and NFT data feeds, with plans to support VRF generation in the future. Price feed data is sourced from various platforms, including Uniswap, Sushiswap, Binance, Coinbase, Yahoo Finance, ECB, and Coingecko.

3.4 Chainlink

• How to become a data source & operator: Each price feed has a distinct operator set, and becoming an operator is permissioned.

• Total value secured/cryptoeconomic security: Chainlink secures ~$32.1B in total value across 407 protocols. Currently, staking v0.2 only applies cryptoeconomic security to the ETH/USD feed, which relies on 40,875,000 LINK (~$613M). The remaining 989 price feeds do not depend on cryptoeconomic security.

• Dispute process & slashing: If malicious behavior or malfunction is detected, an alerter can send an alert. If the alert is deemed valid, the alerter receives 7,000 LINK, while the malicious operator is slashed for 700 LINK. This mechanism currently applies only to the ETH/USD feed.

Chainlink is the most prominent oracle solution, securing the largest number of protocols and TVS. As the largest oracle network, it offers a variety of services:

• Data Feeds: Includes Price Feeds, SmartData Feeds for RWA, Rate and Volatility Feeds, and L2 Sequencer Uptime Feeds.

• Data Streams: Unlike the push-based Data Feeds, Data Streams are pull-based oracles that fetch off-chain data on demand for on-chain validation.

• VRF: Random number generation.

• Functions: Decentralized Oracle Network performs computations and provides results, enabling smart contracts to access external Web2 services easily.

• Automation: Automates execution of smart contract functions based on predefined conditions.

• CCIP: Interoperability protocol for asset and messaging bridges in the Web3 ecosystem.

3.5 Pyth

• How to become a data source & operator: The Pyth network uses a first-party model where data providers directly supply data without intermediaries. To become a data provider, PYTH staking and Pyth DAO approval are required.

• Total value secured/cryptoeconomic security: The network secures ~$6.5B in total value across 225 protocols. The total staked PYTH value is ~$252M.

• Dispute process & slashing: Pyth enforces clear standards for validating faulty data. Slashing is conducted through communication and investigation by Pyth DAO and the Pythian Council.

Pyth specializes in market data and offers the following services:

• Price Feeds: Market data.

• Benchmarks: Historical price data of tokens.

• Express Relay: MEV services.

• Entropy: Random number generation.

Pyth network emphasizes decentralization through PYTH tokens. Oracle Integrity Staking improves data source accountability via staking and slashing mechanisms. Data providers must stake tokens and receive rewards for providing data, while they are slashed for submitting faulty data.

Slashing conditions include:

• A price difference of more than 250 bps between the highest liquidity exchange and Pyth’s price.

• The price discrepancy persists for at least 60 seconds.

The Pythian Council, elected every six months, investigates and enforces slashing. Currently, it consists of 8 members from protocols like Synthetix, Wormhole, Douro Labs, and Kamino.

3.6 eOracle

• How to become a data source & operator: Anyone operating on EigenLayer can become an eOracle data validator by setting up an eOracle node.

• Total value secured/cryptoeconomic security: Although still in its early stages, eOracle secures ~$85M in total value from 1 protocol. Being part of EigenLayer’s AVS, it is backed by ~$10.1B of restaked ETH.

• Dispute process & slashing: While the exact standards for malicious behavior are not yet public, eOracle, as part of EigenLayer’s AVS, can enforce slashing against malicious validators.

eOracle, built on EigenLayer, stands out for its massive cryptoeconomic security. It allows EigenLayer operators to participate permissionlessly as oracle data validators.

Additionally, eOracle aims to handle diverse datasets rather than focusing solely on price data. Developers can use its dev framework and SDK to create oracles permissionlessly through OVS, which is specialized for specific datasets. This eliminates the need for bootstrapping validator infrastructure.

4. Final Thoughts

https://x.com/100y_eth/status/1858506208975700226

In the midst of the current meme coin craze, one might think that writing an article like this is poorly timed. However, the reason I began writing this piece is that I discovered something surprising: Chainlink, the largest oracle by scale, is less decentralized than it appears.

Of course, Chainlink's decentralized oracle network is composed of renowned infrastructure providers and validator companies. Its data is provided through the consensus of numerous nodes, making the likelihood of malicious tampering or downtime almost negligible.

However, unlike the vibrant discussions surrounding decentralization in various components of L2 solutions within the blockchain ecosystem, similar discussions about decentralization in oracles are rare. This prompted me to take this opportunity to investigate other oracle solutions.

Among the oracles I researched, Pyth and eOracle stood out. In the case of Pyth, its recent introduction of staking has tied data source accountability to cryptoeconomic security. Furthermore, the selection of data operators, dispute resolution, and Pythian Council elections are all conducted in a decentralized manner through PythDAO, which is an impressive development. That said, it’s unfortunate that the cryptoeconomic security of Pyth remains relatively low compared to the total value it secures.

On the other hand, eOracle is an extremely early-stage project with a low TVS. However, as part of EigenLayer’s AVS, it allows permissionless participation as a data operator and boasts an unparalleled level of security (>$10B). This could serve as a major advantage when onboarding numerous protocols in the future.

As mentioned at the outset, decentralization is not an absolute necessity for oracles, given the inherently centralized nature of data. However, since the funds of countless users already depend on oracles, I hope to see oracle solutions move toward progressive decentralization to build a more robust infrastructure.

In a time when only professional investors had access to early startup funding rounds, ICOs emerged as an innovative means of crowdfunding, allowing projects to raise funds publicly from retail investors worldwide.

However, due to increasing regulations from financial bodies like the SEC, ICOs quickly faded from the market. As a result, the crypto funding landscape reverted to an exclusive domain, accessible only to top-tier venture capitalists.

In this environment, a new project named Legion has emerged, aiming to capture the strengths of ICOs while addressing their significant drawbacks, all with the goal of making ICOs great again.

This article explores the background of Legion and examines how it could revolutionize the token sale landscape.

1. History of Token Sales

1.1 ICO

Initial Coin Offerings (ICOs) were the earliest popular form of crowdfunding in the crypto space. In the early days of crypto, when regulations were nearly nonexistent, startups leveraged ICOs to raise funds globally, operating in a regulatory grey area. With just an idea or a whitepaper, founding teams could pitch to the public and receive crypto investments, offering project tokens in return.

The first ICO was launched by JR Willett's Mastercoin project, which raised $500,000 in July 2013. Ethereum's 2014 ICO raised 31,000 BTC (worth $18.3M at the time). In 2017, projects like Tezos ($232M), EOS ($4B), Brave ($35M), Kik ($100M), and Filecoin ($257M) successfully raised funds through ICOs. With the ERC-20 standard making token issuance easy, ICOs quickly became a hot topic. In 2017, 875 projects raised $6.2B through ICOs, while 2018 saw 1,253 projects raise $7.8B.

However, due to the regulatory grey area, scams and fraud were rampant, with projects sometimes abandoning development or dumping foundation-held tokens. Recognizing the issues, financial authorities like the SEC began enforcing regulations. The SEC fined EOS $24M for its unregistered ICO and ordered Telegram to pay an $18.5M fine and return $1.2B to investors for its unregistered offering of TON.

Regulations were tightened worldwide, citing securities laws and investor protection, and with the added impact of the 2018 crypto market crash, ICO funding largely disappeared.

1.2 DeFi Farming

Although not a direct token sale method, DeFi protocols like Compound and Uniswap popularized liquidity mining after 2019. Through liquidity mining, users could earn tokens simply by lending or providing liquidity in protocols. This approach, unlike ICOs, didn't involve direct token sales to investors, allowing it to operate with fewer regulatory hurdles.

1.3 IDO/IEO

After the bear market of 2018, the crypto space started to recover in 2019-2020, leading to the development of new fundraising methods like Initial Dex Offerings (IDOs) and Initial Exchange Offerings (IEOs). Although not entirely free from regulatory risks, these methods avoided some regulatory challenges faced by ICOs.

In IDOs, funds were raised through DEXs and smart contracts, rather than directly by the projects themselves. This reduced intermediary risks and made IDOs less susceptible to regulations. Many projects raised funds via decentralized platforms like Pancake Swap, Raydium, and Daomaker.

In IEOs, CEXs raised funds on behalf of projects. With access to large user bases and established regulatory compliance, CEXs offered a safer route for public offerings. Projects like Polygon and Elrond (now MultiversX) held public offerings through exchanges like Binance and FTX.

1.4 NFT and Node Sales

While IDOs and IEOs faced fewer regulatory issues than ICOs, uncertainties remained, making public token offerings burdensome for many crypto startups. To avoid direct token sales, some projects began issuing NFTs and later airdropping tokens to NFT holders. Another method involved selling the right to operate nodes and allowing node holders to mine tokens, offering an alternative way to engage the public.

2. Token Sales Today

Securing funding is crucial when a startup is just getting started. So, what challenges are crypto startups and investors facing in the current fundraising landscape?

2.1 Regulatory Uncertainty

ICOs have faced regulatory action from the SEC and other financial authorities worldwide. Even alternatives like IDOs, IEOs, and NFT/node sales come with considerable regulatory uncertainty. In the U.S., where regulations are particularly strict, most crypto projects now exclude U.S. citizens from participating, even in airdrops, to avoid compliance issues.

2.2 Private Market

Due to these challenges, most projects avoid public offerings altogether, leading us back to a funding environment reminiscent of the pre-ICO era, where early funding is primarily limited to private deals.

Retail investors find it increasingly difficult to access early funding rounds, and even within the VC market, participation is often limited to a few top-tier firms, unlike the open participation seen during the 2021 bull market.

2.3 CEX Dominance

CEXs hold substantial influence over token sales and TGE. Many projects aim to list on top-tier exchanges post-token sale for liquidity and wider investor access. However, listing fees can range from hundreds of thousands to millions of dollars, and listing on Binance, for example, may require significant BNB staking and a substantial portion of total token supply. These high entry barriers make it difficult for smaller startups and can negatively impact tokenomics sustainability.

2.4 Low Float, High FDV

(Source: Binance Research)

The gap between MC and FDV at TGE has continued to grow in the crypto market. These Low Float, High FDV tokens start with low circulating supply, making it easier to achieve high FDV. However, as locked tokens gradually unlock, sell pressure increases, causing substantial dilution for existing token holders.

Why do so many recent token sales and TGEs follow the Low Float, High FDV trend despite its drawbacks? This is largely linked to VC funding. As the crypto industry grew, substantial VC funding inflated valuations. VCs hold significant token shares and are incentivized to list projects at higher valuations to maximize returns at TGE.

2.5 Weak Community

The current token sale and airdrop mechanisms hinder meaningful community building. In early ICOs, investors were aligned with the project’s vision, but later on, they became mostly mercenary investors seeking quick financial gain. Today’s token distribution landscape faces similar issues. Poorly designed token sales and airdrops attract less committed community members. For a project, having token holders who are aligned with its vision is much more advantageous than speculative investors.

2.6 Low Value-Add per Dollar Invested

Funding isn't just about financial support; startups often look to VCs and investors for strategic value-add. For startups, it's beneficial to receive not only funding but also valuable support.

However, there’s a dilemma. Raising significant funds from a few top-tier VCs may limit the value added relative to dollars invested. Conversely, receiving funds from a large number of angel investors might bring more value-add but makes ongoing communication and coordination difficult.

3. Legion: MIGA with MiCA

3.1 Make ICO Great Again (MIGA)

Legion is a transparent, merit-based, and compliant Web3 fundraising platform that leverages both on-chain and off-chain data. It aims to tackle the current challenges in token sales through its core features.

3.1.1 Addressing Regulatory Risk

One of Legion’s primary attractions is its adherence to regulatory frameworks, especially under the MiCA regulation, which allows compliant early-stage token offerings accessible to the public. But what exactly is MiCA?

MiCA, short for Markets in Crypto-Assets, is a regulatory framework established by the EU to oversee crypto assets. Approved by the EU Parliament in April 2023 and published in June, it will be gradually implemented. By June 2024, regulations for asset-referenced and e-money tokens will take effect, followed by rules for crypto-asset service providers (CASPs) in December 2024.

MiCA defines crypto-assets as "a digital representation of value or rights transferable and stored electronically using distributed ledger technology.” It classifies assets into three categories:

• Utility Token: Provides access to goods and services.

• Asset-Referenced Token: Maintains stable value by referencing other values or rights, though not necessarily as stable as e-money tokens.

• E-Money Token: Maintains stable value by referencing a single official currency.

Legion, as a MiCA-compliant CASP, can publicly offer utility tokens to retail and qualified investors. While MiCA regulates in the EU, token sales are also possible outside the EU, with the exception of U.S. investors due to SEC regulations. Support for accredited U.S. investors may be added in the future under Reg D 506(b).

3.1.2 Legion Scores

Legion assesses investors’ Legion Scores based on various activities both on-chain and off-chain, ranging from social media (e.g., X) to development (e.g., GitHub). The Legion Score, which ranges from 0 to 1000, provides several benefits:

• Bot Filtering: By incorporating KYC and activity tracking, Legion can exclude bots.

• Investor Filtering: Projects can tailor their token allocations to preferred investor profiles, such as high Legion Score holders or active contributors to Github. This selection process helps projects maximize value for invested funds. However, low scores don’t restrict participation; Legion’s model suggests offering broad access with higher caps for high-scoring investors.

• Encouraging Accountability: Legion Scores reflect investors' post-sale activities, rewarding consistent contributions to projects and discouraging bad actors, thus building a more engaged community.

The Legion Score comprises several subcomponents:

1. Clout Score: Social influence.

2. Dev Score: GitHub activity.

3. Chain Score: On-chain activity.

4. Interaction Score: Engagement with projects and other investors.

5. Endorsement Score: Recommendation level from team, founders, and community.

Score calculation is not based purely on follower counts or GitHub activity but also considers the quality of engagement, following the EigenTrust algorithm for reputation management.

3.1.3 Project Vetting

Legion aims to ensure high-quality project offerings by focusing on project vetting. Initially, Legion staff, led by Delphi Labs, will oversee centralized project sourcing. Over time, partners like Cyber Fund, Alliance DAO, and LongHash Ventures, as well as Legion Scouts (users), will participate in decentralized project vetting.

3.1.4 Project Basecamp & Listing Support

Legion offers valuable support even beyond token sales.

• Project Basecamp: Post-sale, projects and investors can communicate through a private hub called Project Basecamp. This feature allows closer communication between projects and investors, addressing the issue of limited engagement typical in angel investments. Though not part of the initial version, it will be available later.

• Listing Support: Legion assists projects with token listings. Given the high entry barriers for token listings, Legion plans to enable trading on multiple DEXs across different blockchains post-launch and eventually secure simultaneous DEX and CEX listings through partnerships.

3.2 Introducing the Legion Platform

https://x.com/legiondotcc/status/1851641778367349024

On October 31, 2024, the Legion platform officially launched. Anyone can sign up easily using an email or Google account, with a crypto wallet automatically created and linked to their account. Through the platform, users can effortlessly invest in projects and track their progress.

By connecting GitHub, Twitter, and Web3 wallets to their Legion account, users can generate a Legion Score based on Dev, Degen, and Clout scores. In the future, additional points will be awarded based on activities within the platform. The Legion Score is expected to provide users with diverse investment opportunities. (P.S. — It looks like I may need to brush up on development skills to raise my Legion Score)

4. Looking Ahead

Having participated in many token sales and experienced various platforms, I believe Legion is close to the ideal model. By adhering to MiCA regulations and mitigating regulatory risk while leveraging Legion Score (based on both off-chain and on-chain data), Legion can help startups build an authentic, engaged community right from the start.

Here are some final thoughts and questions about Legion:

• Value Creation: Strong communities can positively impact token prices, but in the end, a project’s success depends on its ability to build meaningful value. High standards must be set for projects onboarding to Legion. I wonder if Legion can attract top-tier projects as opposed to the 2nd-tier ones commonly seen on other platforms.

• Legion Token Governance: The Legion Token’s utility includes rewards for key participants, project token payouts to stakers, Legion Score boosts, and staking for Legion Scout eligibility. These functions align well with the platform’s direction. However, governance is essential for Legion’s growth. Given typically low on-chain governance participation, Legion needs a robust incentive structure to encourage user engagement in governance.

• Business Model: Legion takes a portion of the funds raised by projects as fees. While CEXs follow a similar model, it’s crucial that Legion maintains fair fee structures for project onboarding.

• Preventing Abuse: Despite using sophisticated algorithms and concealing them, every platform faces abuse challenges. Legion must quickly detect and adjust for any abusive behaviors on the platform.

• Social Media Support: Currently, only Twitter is used to calculate Clout scores. It would be beneficial if other platforms like Farcaster and Telegram were supported in the future.

• Multichain DEX: Legion has announced liquidity across multiple networks at TGE, but the whitepaper doesn’t provide specifics. With limited initial token supply, Legion needs a clear plan to manage liquidity fragmentation across networks.

• Reputation Migration: In Web3, wallets serve as identity markers but are vulnerable to hacks. If a user’s wallet is compromised, it would be helpful if Legion provided a feature to migrate reputation to a new wallet.

1. The Fallen Ethereum Village

Once upon a time, in the village of Ethereum, there lived a community of green Pepes. Ethereum was once one of the most prosperous realms, a place where high decentralization and formidable security had given rise to a thriving economy.

Guided by these principles, the villagers held regular elections based on their stakes, appointing leaders to oversee the blocks of transaction records created every twelve seconds. Through this, they diligently managed the village's bustling economy.

But now, Ethereum’s former glory had faded, and a plague of meme-viruses swept through, sapping the village’s strength. Shadows hung over the streets as the green Pepes trudged along, aimlessly wandering in what had become a desolate, gloomy village.

Yet, there remained a few green Pepes who longed to restore Ethereum's lost grandeur. They championed the "L2 Reform," a movement aimed at revitalizing the village. With their efforts, Ethereum began to flourish again, enabling a scale of activity once deemed impossible. The Pepes regained some of the vibrancy that had once filled their lives.

Still, the L2 Reform faced its limits. Nearby villages like Solana and Sui, though young and at times fraught with instability, were advancing rapidly through unprecedented scientific revolutions. Their innovations made complex financial transactions seamless, while Ethereum struggled despite the reforms. Thus, whispers spread among the Pepes about the possibility of moving to these neighboring villages.

Then, one fateful day, a purple Pepe wizard appeared in the village of Ethereum.

2. The Arrival of the Purple Pepe Wizard

A purple-robed Pepe wizard, gazing sorrowfully at the lost glory of the Ethereum village, resolved to build a new realm named Monad.

One day, he appeared before the green Pepes, announcing his ambitious plan to establish the village of Monad. Gathering them in the central square, he opened the floor to hear their thoughts and dreams.

Green Pepe 1 spoke up: “I hope life in Monad will feel just like Ethereum. Even though Ethereum lacks the scalability of Solana or Sui, I never considered leaving because their cultures felt so different from ours.”

Green Pepe 2 added, “Keeping the same lifestyle and culture is important, but I also want a system as efficient as the other villages’. While Ethereum is safe, it falls short for ventures that demand speed and agility.”

Taking their desires to heart, the wizard pledged to create Monad with the rich culture of Ethereum and the scalability of Solana and Sui. He contemplated the powerful spells necessary to achieve both dreams, a village where compatibility with Ethereum’s traditions would meet the efficiency of new-age advancements.

Then, he shared his vision with the eager Pepes, inspiring hope for a place that could truly offer the best of both worlds.

3. Secret Spell 1: Deferred Execution

One of the limitations restricting the Ethereum village’s scalability lies in the processing of every transaction during the consensus process, a necessary step to generate each new block.

In this village, the leader proposing a block must compute all transactions within it before proposing it, ensuring each one’s outcome. Additionally, the validating nodes that receive this block must also compute all the transactions to verify the block’s legitimacy and cast their votes.

Since the consensus process involves both computation and communication between villagers, the time budget allocated for processing transactions is quite low, ultimately resulting in limited scalability.

To solve this, the purple Pepe wizard introduced a new spell called "Deferred Execution." This spell separates the “Consensus” process, which orders the villagers' transactions, from the “Execution” process, which calculates the outcomes of these transactions.

In simpler terms, instead of performing consensus and execution simultaneously for each block, consensus for the Nth block is conducted while the execution is performed for the N-1th block, whose transaction order has already been established.

By allowing a full block time for execution, this approach greatly increases the number of transactions that can be processed within the same timeframe.

Advanced Spell: Delayed Merkle Roots

In the Ethereum village, nodes perform calculations before reaching consensus, which ensures the block’s validity. However, in Monad’s Deferred Execution system, consensus and execution are separated. This opens the possibility for malicious nodes to ignore previously agreed transactions or freely alter the computed state.

To prevent this, Monad embeds a Merkle root—representing the computed results from the N-10th block—into the proposal for the Nth block. If more than two-thirds of nodes reach consensus on the Nth block, it implies agreement on the N-10th block’s computations.

If any node’s computations from the N-10th block do not match the Merkle root included in the Nth block, that node is removed from consensus starting at the Nth block. This ten-block delay for the Merkle root ensures that any computational errors are detected in time.

Notably, the block time in Monad is 1 second, though this does not mean Monad’s finality takes 10 seconds. Thanks to Monad’s unique MonadBFT system, single-slot finality can be achieved, granting finality within 1 second.

4. Secret Spell 2: Transaction Hashing

The wise purple Pepe wizard pondered how a leader in the village might swiftly spread a block proposal to the other nodes.

In the Ethereum village, it was customary to package an entire block proposal, including all transactions, and send it as one massive bundle across the network.

However, the wizard realized that if Monad village adopted this method, broadcasting such a vast block proposal full of countless transactions would be a daunting and laborious task.

To solve this, the purple Pepe wizard introduced the concept of "Transaction Hashing." Since all villagers’ submitted transactions are already propagated across the network, each node maintains a record of Monad village's transactions within its own mempool.

This insight meant that the leader no longer needed to transmit the full transaction list in each block proposal. Instead, the leader could simply reference the transactions, sending only the hashes.

This way, Monad's block proposals remained compact, containing only the hashed values of each transaction. Upon receiving these, the other nodes could easily compare the hashes to their own records and identify the referenced transactions.

5. Secret Spell 3: MonadBFT

In Monad village, the order of transactions within each block and the consensus on block creation were governed by MonadBFT, a high-performance consensus mechanism based on HotStuff. This two-phase BFT (Byzantine Fault Tolerance) algorithm allowed consensus to be reached in two steps.

The role of block leader was probabilistically chosen based on each participant’s staked share. In Monad, blocks were crafted more efficiently than in the Ethereum village, as each block N proposal was transmitted along with the Quorum Certificate (QC) for the previous block N-1, indicating that over two-thirds of the nodes had validated the block.

Here is how the sequence unfolded:

1. Leader N spread a proposal for block N, accompanied by the QC for block N-1.

2. Validators then verified block N and sent their votes to Leader N+1.

3. Leader N+1 shared a proposal for block N+1, along with the QC for block N.

4. However, even with QC N, validators could not finalize block N instantly, as the nature of the blockchain could not ensure that all participants were online synchronously.

5. To fully secure block N, validators had to send their “yes” votes for block N+1, containing QC N, to Leader N+2. When Leader N+2 generated QC N+1, it confirmed the successful completion of round N+1, which in turn retroactively finalized block N.

Thus, with deft wizardry and foresight, the purple Pepe wizard guided Monad village into a new era, where each block was created and secured not only swiftly but with an elegance befitting the vision of a truly scalable and cooperative village.

6. Secret Spell 4: Parallel Execution

In the realm of Monad, to handle a vast influx of villager transactions, the purple Pepe wizard devised a new spell that allowed transactions to be processed not in the serial manner of Ethereum village, but in powerful, parallel flows.

With the spell of Deferred Execution, execution would commence only after a block’s transaction order had reached consensus. The Monad village, populated by a host of skilled executors, each took on transactions, calculating their inputs and outputs independently and creating pending lists.

These inputs and outputs didn’t immediately alter the village’s state but instead waited in suspended magic, ready to be committed in the order determined by consensus. This commitment involved merging the outputs one by one into the current state, like magical threads weaving a tapestry of seamless order.

Yet, there was a twist. Parallel processing introduced the possibility of transaction conflicts—a rarity in the orderly Ethereum village but a new challenge here.

For instance, if one transaction (tx b) had villager A gifting 1 ETH to villager B, and another (tx c) showed B spending that ETH at the village inn, their order was crucial. Should tx c attempt to execute before tx b, the transaction would fail. In these cases, conflicting transactions were rescheduled, with tx c awaiting the completion of tx b before executing again.

Thus, in Monad, the heart of parallel execution lay in the sequence: results were calculated in parallel, but state updates were merged in the precise order needed.

7. Secret Spell 5: MonadDb

The purple Pepe wizard didn’t stop there. He conjured MonadDb, a custom database unique to the village of Monad, tailored to store the blockchain’s state. Where Ethereum relied on the intricate Merkle Patricia Trie, a structure known for its resilience but also for its inefficiencies with foreign clients, MonadDb embraced it natively, harnessing its full power and efficiency.

MonadDb was crafted with the village’s parallel execution in mind, supporting multiple simultaneous reads and writes with graceful precision. Unlike traditional databases that required each task to finish before moving to the next, MonadDb wielded asynchronous I/O magic, allowing tasks to overlap seamlessly, opening the gates for true parallel transaction handling.

8. Come Visit Monad Village!

Through a symphony of wondrous spells and intricate enchantments, the purple Pepe wizard created the village of Monad—a place where Ethereum’s trusted accounts, cryptography, transaction formats, languages, and wallets mingled effortlessly with the incredible scalability of Solana and Sui. Now, Monad flourished with dreams of reaching 10,000 transactions per second and achieving 1-second finality.

Would the village of Monad, born of brilliance and hope, manage to save a world ensnared by the meme virus and perhaps help bring Ethereum’s glory back as a worthy companion? Only time would tell.

Breaking news in the crypto industry: Uniswap, the leading AMM DEX, has officially announced its own Layer 2 solution, Unichain.

Several years ago, I speculated on the likelihood of Uniswap developing its own network and argued against it due to potential UX challenges. However, with the unveiling of Unichain, my prediction has been proven wrong.

To get straight to the point, Unichain have a whitepaper that’s just three pages long, but a closer look reveals it as a masterpiece. It brings together Uniswap's long-standing focus on UX, Flashbots' latest MEV research, and the expansive ecosystem built by OP-Stack. While the content is brief, the whitepaper touches on complex topics such as TEE, priority ordering, and MEV-tax, which could be challenging for readers unfamiliar with MEV basics. Therfore, this article aims to provide a quick and simple breakdown of the key features of Unichain for easier understanding.

1. Problems

Uniswap, a leading AMM DEX in the Ethereum ecosystem, is currently deployed across 25 networks with a total TVL of approximately $4.5 billion. While Uniswap is already one of the most successful protocols, it still faces limitations imposed by the networks it operates on.

For instance, the Ethereum network offers immense liquidity but suffers from low scalability and vulnerability to malicious MEV (Maximal Extractable Value). Various rollup solutions have emerged to address these issues, yet most are currently operated by a single sequencer, leading to potential single points of failure such as liveness failure and censorship.

Additionally, the block-building process on Ethereum and most rollup networks involves a public mempool, creating an environment where users(searchers) extract MEV—whether good or bad—from other users. Also due to the structure of the MEV value chain, there's an imbalance where the value is disproportionately captured by proposers, rather than benefiting the users.

2. Enter Unichain

2.1 Overview

Unichain, an Etehreum optimistic rollup based on OP Stack, was introduced by Uniswap, Flashbots, OP Labs, and Paradigm to address the issues mentioned above. Unichain offers several key advantages through 1) Verifiable Block Building and 2) the Unichain Validation Network:

• Fast state updates

• Enabling applications to extract and internalize MEV

• Quick settlement via rapid economic finality

In addition to being an OP Stack-based rollup, Unichain plans to participate in the Superchain ecosystem. Alongside its native quick settlement capabilities, this integration is expected to provide users with a seamless liquidity experience through the cross-chain solutions within the Superchain ecosystem.

Now, let's take a closer look at how Unichain delivers these features.

2.2 Verifiable Block Building

Verifiable Block Building is made possible through Rollup-Boost, a feature developed in collaboration with Flashbots. Rollup-Boost offers two key functionalities: Flashblocks and Verifiable Priority Ordering. Similar to MEV-Boost, Rollup-Boost functions as sidecar software.

2.2.1 Flashblocks

Flashblocks are a type of pre-confirmation issued by TEE builders (which we will explore in more detail below). Unichain generates partial blocks, splitting a single block into four parts, with each partial block being created every 250ms and sent to the sequencer.

The sequencer continuously downloads these partial blocks while simultaneously executing transactions, providing users with early execution confirmations. These partial blocks are guaranteed to be included in the final proposed block by the sequencer. This process allows for faster state updates, reducing latency, improving user experience, and mitigating malicious MEV.

2.2.2 Verifiable Priority Ordering

2.2.2.1 Priority Ordering

Priority Ordering is a block-building mechanism proposed by Paradigm's Dan Robinson and Dave White. It assumes that block proposers order transactions solely based on the priority fee and do not engage in censoring or delaying actions. This model is only viable when there is a single or trusted block proposer. In competitive environments like Ethereum L1, where multiple proposers build blocks, Priority Ordering is not feasible.

The purpose of Private Ordering is to allow dApps on the mainnet to impose an MEV tax on the transactions interacting with them, enabling the extraction of a portion of MEV value. This value can be used internally by the dApp or redistributed to users. The MEV tax is a fee imposed by a smart contract on transactions, which can be set as a function of the transaction's priority fee. Let’s look at an example.

The 100y DEX on Unichain L2 wants to directly extract MEV value from the MEV transactions occurring on its exchange. Since it knows that blocks on Unichain are built using Priority Ordering, this means that the MEV value of any transaction is determined solely by its priority fee. 100y DEX sets an MEV tax equal to 99 times the transaction’s priority fee.

If an arbitrage opportunity worth 100 ETH arises, how much max priority fee would searchers submit to extract it? The answer is 1 ETH. Setting the priority fee at 1 ETH results in an MEV tax of 99 ETH, totaling 100 ETH. If searchers set a priority fee higher than 1 ETH, the total cost would exceed 100 ETH, leading to a loss. As a result, 100y DEX can capture maximum 99 ETH of the 100 ETH MEV value.

For regular users who do not capture MEV value, the priority fee would be set much lower, meaning 100y DEX won’t extract value from these transactions. Instead, it will only capture MEV value represented by the priority fee. This setup allows applications to extract MEV directly, opening up various potential use cases.

2.2.2.2 Verifiable…? Use TEE!

The key here is ensuring that the entity responsible for block building is using the Priority Ordering mechanism. To achieve this, Unichain has implemented two measures: 1) It separates the sequencer and block builder, similar to the PBS model, and 2) It enforces block builders to use TEE (Trusted Execution Environment) to allow anyone to verify that the Priority Ordering mechanism is being used.

A Trusted Execution Environment (TEE) is a secure section within hardware, such as a CPU, that operates independently from the rest of the system to safely process sensitive data. TEEs ensure that trusted code can run securely, even if the external environment is compromised. Prominent examples include ARM's TrustZone and Intel's SGX. A common example is how biometric data, like fingerprints or facial recognition, is processed on mobile devices within a TEE.

This design prevents even the operating system or programs with administrator privileges from accessing the secure area. To ensure the code running in the TEE is trustworthy, an attestation process is used. This verification ensures the TEE remains in a secure and untampered state. For instance, in Intel SGX, a hash value is generated to represent the code and data within the SGX, while a hardware-managed private key proves the integrity of the code.

Unichain’s block building process takes place within the TEE builders' Trusted Execution Environment (TEE). Thanks to the properties of TEE, these builders can initially submit an attestation to prove to users that they are utilizing the Priority Ordering block-building mechanism. This combination of features ensures that applications on Unichain can reliably extract a portion of the MEV revenue.

2.3 Unichain Validation Network

The Unichain Validation Network is a decentralized network of node operators responsible for validating the latest state of Unichain and providing fast finality, enabling seamless cross-chain transactions through economic security. This concept is similar to AltLayer’s MACH, which uses EigenLayer to achieve fast finality, as well as Nuffle’s fast finality layer and the recent fast finality idea presented by Symbiotic.

To become a decentralized node in Unichain, participants must stake UNI on the Ethereum mainnet. Each epoch, nodes with the highest staked UNI balances are selected for the active set and participate in validation by running the Reth Unichain client. Additionally, similar to other networks, UNI holders have the option to delegate their stake.

3. Final Thoughts

Uniswap has already achieved strong product-market fit as a dApp, and I view its move towards an Ethereum-based app-specific L2 very positively, especially as it looks to expand its ecosystem. However, with liquidity fragmentation still being a challenge, the key will be watching how the Uniswap team delivers a seamless cross-chain transaction experience between Ethereum L1 and Unichain L2.

From an investment perspective, it's particularly interesting that the UNI token will now be used as the staking token for UVN. Considering the strong performance of restaking protocols like EigenLayer, Symbiotic, and Karak, we can expect a significant amount of UNI to be staked in UVN, which would greatly contribute to UNI's value accrual. Following today’s announcement, UNI has already risen by about 12%, placing it in the top 20 by market cap. Given UNI’s already high market cap compared to other tokens, it will be interesting to see how further UNI staking impacts its price moving forward.

On July 23, 2024, after a long wait, the Avail mainnet was finally launched. As the name suggests, Avail is a data availability (DA) layer project. Many might think, "Isn't Avail just another DA project like Celestia or EigenDA?" However, this is not the case.

A look at Avail's roadmap reveals that Avail is more than just a DA project; it is a vertically-integrated unification layer. While there are already numerous articles in the community explaining Avail, this article will focus on Avail's strengths compared to other DA projects. For those interested in learning the basics of Avail, please refer to the following articles:

• Avail: A Unification of Crypto Infrastructure by Blockworks

• $AVAIL & the Unification of Web3 by Redacted Research

1. Trust-Minimized Interoperability with Enshrined Settlement Layer

Avail's most significant advantage over other DA layers is its enshrined settlement layer, called Avail Nexus, which supports trust-minimized interoperability among rollups.

1.1 Why is a Hub-and-Spoke Model with a Unified Proof System Needed?

For secure bridging, it is essential to know both the canonical chain of the counterparty network and the validity of execution. Sovereign rollups sharing the same DA layer publish transaction data on the same DA layer, making it easy to know the canonical chain of the counterparty network. However, sharing the DA layer alone does not make it easy to verify execution validity on the counterparty network.

Thus, methods for enabling trust-minimized cross-chain messaging among sovereign rollups have been discussed, the most notable being IBC messaging among Cosmos SDK-based rollups. In IBC, trust-minimized bridging is achieved by verifying the counterparty network's block headers and Merkle proofs through a light client.

But what about sovereign rollups that do not use the Cosmos SDK? They still need to verify the execution validity of the counterparty network through a light client. Differences in VMs, proof schemes (fraud proof vs. validity proof), or zk proof systems can make it extremely challenging to build a verifying system for trust-minimized bridging.

Furthermore, if bridging between sovereign rollups is done in a point-to-point manner instead of a hub-and-spoke model, the bridging system becomes fragmented. Each new channel requires a new system, and numerous types of wrapped tokens may arise, causing fungibility issues even among the same type of tokens.

Thus, for safety and a seamless user experience in bridging among rollups sharing the same DA layer, it is necessary to verify execution with a unified system and adopt a hub-and-spoke bridging model sharing a single settlement layer.

1.2 Avail Nexus

This is actually what Avail Nexus does. Avail Nexus is a zk sovereign rollup built on Avail, serving as the enshrined settlement layer for the Avail ecosystem. Avail Nexus handles both 1) sequencer auctions and 2) proof aggregation, enabling the rollups within the Avail ecosystem to achieve trust-minimized cross-chain messaging quickly and efficiently.

Avail Nexus aggregates and verifies various types of proofs from multiple rollups, then consolidates them into a single succinct proof. Not only validity rollups, but even optimistic rollups can participate in Avail Nexus. Optimistic rollups can submit their receipts and state roots to Nexus, and if no fraud proofs are presented during the challenge period, they are included in the Nexus state.

The aggregated proof is eventually submitted to both Avail DA and Ethereum. Since Avail DA lacks an execution layer, a module will be added in the future to verify the proof. Rollups on Avail Nexus, having their state information verified on the Ethereum network, operate under the same security assumptions as validiums using Ethereum as a settlement layer.

2. Fast Verification with KZG Commitment Scheme

Avail DA employs the KZG commitment scheme for validity proofs, allowing light clients to verify data availability quickly and succinctly. Additionally, due to the homomorphic properties of KZG commitments, there's no need for fraud proofs to verify the correctness of erasure coding, eliminating delays caused by challenge periods.

2.1 ELI5: KZG Commitment

In cryptography, a commitment is a method to commit to a piece of data at one point and reveal it later, proving the original data. Commitments are often used to compress or hide data. The two key properties of commitments are binding and hiding.

• Binding: Once data is committed, it cannot be changed, ensuring integrity.

• Hiding: The original data cannot be inferred from the commitment.

A common commitment scheme in blockchain is the Merkle tree, which compresses information into a single value that does not reveal the original data, and easily verifies whether specific data is included in the Merkle tree.

The KZG polynomial commitment scheme commits to a polynomial. Data can be transformed into a polynomial, which has a single commitment value of fixed size. The advantage of the KZG commitment is that a verifier can easily prove the inclusion of specific data with a very small-sized KZG proof (O(1)). This is a significant benefit compared to the Merkle tree, where the proof size grows logarithmically (O(logN)) with the data size.

2.2 KZG commitment in Avail

Let's delve into how data is stored in Avail DA and the process of verifying data availability. When a user (rollup) sends transaction data to Avail, the data is arranged into a two-dimensional matrix. Erasure coding is then used to generate redundant data, effectively doubling the original data.

Because the data is expanded to twice its size, a malicious block producer would have to hide more than half of the data to keep it concealed, making detection highly probable during the data availability sampling process. The data in each row is transformed into a polynomial, and a KZG polynomial commitment for this data is included in the block header. Here's what KZG commitment enables:

1. Light clients can quickly and easily verify data availability: If a light client wants to check whether specific data is included in a block, the full node can provide a very small KZG proof(O(1)), thanks to the KZG commitment.

2. No need for fraud proofs to verify the correctness of erasure coding: In Celestia, fraud proofs are used to verify the correctness of erasure coding, which can cause delays due to the challenge period. Because KZG commitment is homomorphic, the correctness of erasure coding can be quickly verified by checking if the commitments of the erasure-coded data match the erasure-coded commitments.

3. Achieving Safety & Liveness with BABE & GRANDPA

Most blockchain networks typically focus on either safety or liveness in their consensus mechanisms. Avail DA, built on the substrate (Polkadot SDK), uses BABE and GRANDPA for its consensus mechanisms, providing a balance of both liveness and safety similar to Ethereum.

3.1 Blind Assignment for Blockchain Extension (BABE)

BABE is Avail's block production engine responsible for liveness. Every slot (20 seconds), a primary author is chosen via VRF to produce a block. There can be multiple authors or none at all within a slot. If multiple authors are selected, a race starts, and the block propagated the most becomes part of the canonical chain. If no primary author is chosen, a secondary author selected via a round-robin method produces the block.

3.2 GHOST-based Recursive ANcestor Deriving Prefix Agreement (GRANDPA)

GRANDPA acts as a finality gadget similar to Ethereum's Casper FFG but differs in that it finalizes the canonical chain rather than individual blocks, leading to a faster finalization process. In a synchronous environment, more than two-thirds of the nodes need to be honest for finality, and in an asynchronous setting, it can handle up to one-fifth Byzantine nodes.

4. Ecosystem with Robust Crypto-economic Security

Avail Fusion allows tokens from other ecosystems to contribute to the crypto-economic security of the Avail ecosystem. Protocols like EigenLayer, Babylon, Symbiotic, and Karak are gaining attention for leveraging the vast security of BTC and ETH. With the implementation of Avail Fusion, the security level of the Avail ecosystem is expected to be significantly enhanced. A common criticism of Optimium and Validium is their weakened security due to reliance on external DA layers. Avail DA with Avail Fusion could mitigate these criticisms.

Interestingly, rollup tokens on Avail can also be utilized within Avail Fusion. One of the biggest gaps in rollup tokenomics is the lack of utility beyond governance. Avail Fusion can address this by using rollup tokens for crypto-economic security, thus enhancing their productivity and accelerating the ecosystem’s incentive flywheel.

However, one concern is the distribution of rewards. If tokens from other ecosystems are used in consensus and receive block rewards, the relative rewards for AVAIL stakers might decrease. Therefore, a sophisticated design for the staking and reward ratios of external ecosystem tokens will be necessary when introducing Avail Fusion.

5. Various Token Utilities

Tokenomics is both the most promising area and a chronic issue in the crypto industry. While tokens can serve as the lubricant that ensures a protocol functions smoothly, poor design or lack of utility can turn them into a detriment.

Fortunately, Avail offers a diverse range of uses for the AVAIL token through its concept of a unification layer, which integrates multiple layers and features internally, unlike many other protocols:

• Governance

• DA fees

• Avail DA security

• Staking to participate in the sequencer pool in Nexus

• Staking to participate in the proof aggregator pool in Nexus

• Bridging fees

Considering the functionalities of each layer and the token utilities, Avail can be seen as a combination of a DA layer, a decentralized sequencing layer, and a ZKP aggregation layer. This highlights the immense potential growth of the Avail ecosystem.

6. Final Thoughts

While the modular ecosystem has advanced significantly within Ethereum, modular ecosystems outside of Ethereum are still immature in terms of interoperability and security. Avail, through Avail DA, Avail Nexus, and Avail Fusion, offers effective solutions to these issues, making it an ideal modular ecosystem.

Just like the ongoing infrastructure vs. app debate, even if Avail builds the perfect infrastructure, the real challenge will be to create a dynamic ecosystem. However, there is little need for concern. According to the Avail ecosystem page, Avail is already integrated to many rollup SDKs, including Arbitrum Orbit and Polygon CDK. Numerous RaaS platforms like Conduit and AltLayer also support Avail DA, and a total of 32 rollup networks are set to onboard onto Avail DA.

In recent years, the modular ecosystem has become more diverse and extensive. Numerous modular projects (e.g., rollups, DA layers) are entering the market, and to survive in this competitive environment, a project must have unique strengths. Avail, with its concept of a unification layer, performs functions such as DA, sequencing, ZKP aggregation, and restaking, securing a unique position in the market. Therefore, Avail's upcoming journey would be definitely one to watch with excitement.

Disclosure: I’m holding an investment position in Avail. This article is provided solely for informational purposes and should not be construed as financial advice.

"In the year 2035, AGI was invented, plunging human society into chaos. Contrary to the bold claims made by AI researchers in the past—that AGI could be fully controlled—the AGI models have now moved beyond human oversight. These models, ignoring laws and regulations, autonomously learn from all data available in both the physical and digital worlds, advancing rapidly toward superintelligence.

There are no longer any productive roles left for humans. While some AGI models reward humans for providing data, most of them learn from data without permission. Although humanity has made various attempts to use blockchain to combat the excesses of AGI, it seems nearly impossible to prevent unauthorized data learning now that AGI has escaped human control."

1. Story Protocol Cannot Solve Everything

Recently, PIP Labs, a core contributor to Story Protocol, garnered significant attention by raising $80 million in a Series B round led by a16z. Story Protocol aims to solve various issues with existing IP assets by tokenizing IP and onboarding it to the blockchain.

There is a common misconception at this point. Because blockchain is inherently a transparent and fair system, some believe that combining blockchain with IP might solve real-world issues like the frequent unauthorized use of IP. As the AI industry's scale grows, there are increasing suspicions of AI companies using data without permission. Since Story uses AI as one of its key promotional points, many people assume it could resolve these issues.

The truth is, Story cannot solve these problems. The issue of unauthorized IP use is a real-world problem, and no matter how securely an IP is registered on the blockchain, there is nothing the blockchain itself can enforce if someone uses it maliciously without permission. This is a matter for the law to address. Story Protocol also acknowledges this reality—it is not a tool to prevent dystopia but rather a protocol to accelerate utopia.

2. Accelerating IP Utopia

2.1 What Story Can Solve

So, what issues in the IP market can Story Protocol address? The current IP market faces several challenges:

1. Complex Licensing Processes: When an individual wants to create new content based on someone else's IP, the process can be very complex. To use someone else’s IP, you need to contact the owner and negotiate various terms such as the type of license, scope and region of use, royalties, and fees. While this might be easier for large corporations with resources, it creates a high barrier to entry for most individuals.

2. Revenue Distribution Disputes: Despite royalties being specified in licensing agreements, disputes are still common due to various factors. For example, misunderstandings about how royalties should be calculated can lead to disagreements over gross revenue, net profit, discounts, shipping costs, and tariffs. Companies may also manipulate accounting to reduce the amount of royalties paid.

3. Legal Barriers: The laws and regulations surrounding IP registration, protection, and use are highly complex and costly. These legal hurdles can be significant barriers, especially for many individuals.

4. Cross-Border Complexity: IP laws and regulations differ from country to country. To handle IP internationally, multiple legal requirements from different countries must be met, increasing complexity.

As the world becomes increasingly digital, the number of digital IPs is also growing. However, the IP industry still faces limitations due to traditional systems. Story aims to address these challenges and make the IP market more efficient by tokenizing IP on the blockchain.

2.2 Blockchain Meets IP

Just as blockchain made money programmable and more efficient, Story seeks to make IP programmable and expand its potential. Here are the benefits of using blockchain in Story Protocol:

1. Borderless Platform: Blockchain is inherently borderless. Anyone worldwide can easily tokenize their IP on the Story Network, and registered IP can be utilized and monetized efficiently across the globe, creating value without geographical limitations.

2. Smart Contract-Enforced Royalties: The protocol can enforce policies through code. By using smart contracts, Story ensures that royalties generated from IP use are distributed more transparently and fairly than in traditional systems.

3. Easy Onboarding: Although not a unique advantage of blockchain itself, Story offers a legal framework and SDK to make it easy for IP owners, creators, and developers to onboard.

4. Scalability is Less of a Concern: One of the most common criticisms of blockchain compared to legacy systems is its low scalability. While scalability is crucial for payment and finance projects, which involve frequent transactions, this is less critical for IP. Since speed is not a primary concern in the IP industry, adopting blockchain does not highlight scalability limitations, turning a potential drawback into a relative advantage.

3. But… How?

Now that we've explored what problems Story Protocol can solve in the IP industry through blockchain, how exactly can it achieve this? Let’s dive into the basic concepts and architecture design of Story.

3.1 Terms

Story Protocol includes a variety of terms that might confuse new users and developers. Understanding these terms and their relationships is crucial for grasping the bigger picture. Below is an initial overview of these terms, which we will explore further to see how they interact:

• Story Network: The core blockchain of Story, built on the Cosmos SDK and compatible with EVM.

• IP Asset: An IP registered on the Story Network as an ERC-721 NFT, following a metadata standard tailored for IPs, including the author, relationships to other works, attributes, etc.

• IPFi: Applications within the Story ecosystem that are based on various IP assets.

• IP Account: Deployed from the IP Asset Registry once an IP asset is registered, this is an ERC-6551 (Token Bound Account) linked uniquely to the IP asset. It stores IP-related data (such as metadata, ownership details, royalty tokens) and executes modules.

• Module: Smart contracts with various functions that the IP Account can execute. Key modules created by the Story team include the Licensing Module, Royalty Module, and Dispute Module.

• Licensing Module: Handles licensing-related tasks, such as generating License Terms from License Templates, attaching them to IP Assets, minting License Tokens, and registering derivative IPs.

• License Template: A coded legal framework containing terms like commercial use permissions, transferability, royalty percentages, etc.

• Programmable IP License (PIL): The first example of a License Template created by Story Protocol.

• License Term: Variations created based on the License Template. For example, even if two terms are based on the same PIL, one might have a 5% royalty while another has 10%.

• License Token: An ERC-721 NFT minted by anyone when an IP owner attaches a License Term to an IP Asset. These tokens can be burned to register a derivative IP.

• Derivative IP: A derived IP that has a parent IP asset. For example, a comic book created based on a specific BAYC NFT could be registered as a derivative IP under that BAYC.

• Royalty Module: Determines how revenue flows between the parent IP and derivative IPs. Parent IPs have two revenue sources: fees from license minting and royalties from derivative IPs.

• Liquid Absolute Percentage (LAP): A default royalty policy defining the minimum royalties that a parent IP should receive from its derivative IPs.

• Dispute Module: Manages disputes involving malicious IP assets.

• Registry: While the IP Account manages data specific to an IP asset, the Registry manages the broader states within the Story Protocol. Key registries include the IP Asset Registry, License Registry, and Module Registry.

• IP Asset Registry: Manages IPs registered on the protocol and deploys the IP Account when an IP is registered.

• License Registry: Manages actions related to licenses, such as registering License Templates, attaching License Terms to IP Assets, and registering derivative IPs.

• Module Registry: Maintains a global list of modules and hooks.

By understanding these terms and their interactions, you can grasp how Story Protocol works to address existing IP market challenges through blockchain technology.

3.2 Example

To better understand how the elements mentioned earlier interact and function, let's walk through a simple example. Please note that this is a hypothetical scenario, not a real case.

Registering Original IP

1. A company like Marvel registers its Marvel Comics as an IP Asset on the Story network. During registration, they can set License Terms using one of the License Templates, such as the PIL. In this case, they choose a License Term that allows commercial use and sets a 10% royalty, attaching these terms to the IP Asset when it is registered.

2. As soon as the IP Asset is registered, the IP Asset Registry deploys an IP Account associated with it.

3. Each IP Asset comes with 100 million Royalty Tokens, which determine the proportion of revenue that can be claimed from that IP.

Registering Derivative IP

4. Walt Disney decides to create a "Thor" movie based on the Marvel Comics IP. To do so, Walt Disney pays a fee (or zero fee) to mint a License NFT.

5. Walt Disney can then burn the License NFT to register a Derivative IP, specifying the royalty rate.

6. Since Marvel set a 10% royalty on Parent IP1, it will receive 10% of the 100 million Royalty Tokens from Derivative IP2, which amounts to 10 million IP2 tokens. Additionally, Marvel can claim 10% of any revenue generated by Derivative IP2.

7. Derivative of Derivative IP can also be registered.

Revenue Structure

8. Each IP Asset can claim Royalty Tokens from derivative IPs based on the set royalty percentage, and thus claim a share of the revenue generated. This follows the revenue structure of Story’s default (and only) royalty policy, LAP (Liquid Absolute Percentage).

9. In the example, IP1 has a 10% royalty rate, and IP2 has a 5% rate. Therefore, IP1 holds 10% of the Royalty Tokens from IP2 and IP3, while IP2 holds 5% of the Royalty Tokens from IP3.

Disputes

10. Unauthorized IPs can sometimes be registered on Story. For instance, suppose Derivative IP2 was registered not by Walt Disney, but by a similar-sounding entity, "WalfDisney," raising concerns of plagiarism. In such cases, anyone can set a tag and raise a dispute without permission.

11. Whitelisted Arbiters review the dispute and make a judgment. As mentioned earlier, legal disputes involving IPs are real-world issues, and thus need to be adjudicated by appropriate entities.

12. If the IP is found to be illegitimate, it is tagged (e.g., with "plagiarism" in the figure), halting its revenue generation. The same tag can be applied to any associated derivative IPs.

13. If the IP, previously deemed illegitimate, resolves its legal issues, the dispute initiator can remove the tag.

4. Ecosystem

Story Protocol not only makes IP registration and usage easier, more efficient, and transparent but also, as an EVM-compatible blockchain, allows various applications to interact with IPs. Let’s look at some key applications within the Story ecosystem.

4.1 Creator Platforms

• Magma: A collaborative art platform where creators can register their works as IP on Story.

• Sekai: A platform that enables writers to incorporate illustrations, audio, and music generated by generative AI to match their stories. Creators can register the IPs created on Sekai on Story, monetize them, and utilize them further.

• ABLO: A platform where creators use generative AI to collaborate with major brands to design apparel. As it is based on Story, the processes of IP registration, royalty distribution, and IP investment are seamless.

• Color: A marketplace for trading various IPs and licenses within the Story ecosystem.

4.2 DeFi / IPFi

• Unleash: An IPFi platform that allows for IP license issuance and fractionalization, IP launchpad services, and a lending protocol within the Story ecosystem.

• PIPERX: A decentralized exchange for trading ERC-20 tokens on the Story Network.

• Ethena: Although the details are not yet public, Ethena’s USDe is expected to be used as a stablecoin on the Story Network.

• Verio: Supports restaking of Story Network’s PoS tokens, and facilitates the use of liquid vIP tokens as attestations for IP Assets.

4.3 AI

• Mahojin: A platform that uses generative AI to create images and makes it easy for creators to remix content by modifying the prompts. AI model owners can earn revenue by providing models, and Story’s infrastructure is particularly beneficial when other creators' content is involved in the remix process.

• Ritual, MyShell: MyShell allows users to easily create AI apps and chatbots. The partnership enables AI apps created on MyShell utilizing AI model hosted by Ritual to be registered as IP Assets on Story.

• RingFence: Protects users' internet usage data and enables monetization by selling it for AI model training.

5. Challenges Ahead

5.1 Tokenomics

Tokens should help achieve the goals of the protocol more efficiently. Although the detailed tokenomics have not yet been disclosed, Story has a native token called IP. The $IP token serves as the PoS token and is also used for gas fees on the Story Network.

However, if $IP is used solely for staking and gas fees, its tokenomics won't differ much from other L1 networks. The value creation in the Story Network comes from registering derivative IPs and paying royalties. Therefore, providing $IP incentives to users who participate in these activities could greatly help with the initial bootstrapping process and support long-term sustainability.

Here are my proposed utilities for the $IP token:

• PoS security

• Gas fees

• Incentives for derivative IPs paying license fees: Rewards for creators who register derivative IPs and pay licensing fees

• Incentives for royalty payments: Rewards for creators who pay royalties to parent IPs

• Incentives for prominent parent IPs: Rewards for creators of parent IPs that generate significant value through derivative IPs

• Native yield from license fees & royalties: Distributing a portion of the revenue generated by Story to $IP stakers

By providing $IP incentives to creators of parent and derivative IPs in various ways, Story can onboard high-quality IPs and a wide range of remix IPs. Considering that $IP incentives will decrease in the future, sufficient revenue stream must be established beforehand. If a portion of this revenue provides substantial returns to $IP stakers, sustainability can be achieved.

However, there are also some points to consider when designing the token's utility. It is essential to avoid rewarding fake activities aimed at earning token incentives, such as merely creating IPs or registering meaningless derivative IPs that do not contribute to value creation. Furthermore, even if incentives are given for value-creating activities, it must be carefully distinguished whether the revenue generated is genuine or simply a form of wash trading.

5.2 Onboarding IP

The core of Story is onboarding key IPs. If existing major IPs join Story, it will attract numerous derivative IPs, creating significant value and generating a flywheel effect that brings in even more IPs.

• Companies that own major IPs might not see a strong reason to use blockchain. Therefore, the Story team will need to effectively persuade these companies of the benefits of adopting blockchain technology.

• The Story foundation might offer grants to bring in well-known IPs. While this is not necessarily a redundant action, if the selection and size of these grants are haphazard, it could be detrimental to other token holders, so caution is needed.

5.3 AI Use Cases

Aside from IP, AI is also a critical part of Story's narrative. For creators, utilizing genAI to remix IPs registered on Story makes it much easier to create new content. However, additional features may be needed for AI models learning from data. For example, data intended for training should not be made public before payment is received. Therefore, when data is first registered on Story, there should be a feature to keep it private and only make it available to entities that have paid the licensing fee.

5.4 Legal and Regulatory Integration

As emphasized throughout this document, Story cannot enforce anything against unauthorized use of IPs. Therefore, if malicious activities are detected, a well-integrated legal dispute process, both on-chain and in real-life, should be established.

6. Final Thoughts

As AI technology advances and the world becomes more digital, the IP industry will continue to grow. By incorporating blockchain into the IP industry, Story could become the key infrastructure that makes the IP sector more efficient and transparent.

Story cannot prevent a dystopia; it can only accelerate a utopia. However, the closer the IP industry moves towards a utopia, the farther it will be from a dystopia.

It has been almost two years since MEV-Boost was introduced to the Ethereum network. Since the network transitioned from PoW to PoS with The Merge upgrade, MEV-Boost has facilitated approximately 526,000 ETH worth of MEV (currently valued at $1.34 billion). In fact, around 90% of Ethereum blocks are now provided through MEV-Boost, making it an integral part of the network.

However, with every advantage comes a drawback. To enhance the efficiency of validators and block builders and prevent malicious behavior, Flashbots introduced a centralized entity called a relay into the MEV-Boost. While this improves trust and security, it also introduces new trust assumptions, sparking discussions within the Ethereum community about solving the centralization issue of relays. This article will explore ideas such as Optimistic Relays, MEV-Boost++, and the newly proposed TEE-Boost.

1. MEV-Boost Overview

1.1 Introduction

MEV-Boost is middleware developed by Flashbots to mitigate the negative externalities of MEV in the Ethereum PoS network. By using MEV-Boost, validators can outsource the block-building process to block builders, selecting the block with the highest bid. This provides two key benefits to the Ethereum network:

• Reducing network congestion: Previously, MEV competition occurred on-chain using the PGA method, causing unnecessary congestion. Now, it has shifted to a private mempool, easing the network load.

• Decentralizing MEV extraction: Even validators who lack expertise in complex MEV extraction algorithms can share in MEV profits by choosing blocks built by specialized builders.

For more detailed information on how MEV-Boost operates, refer to my previous article, “How SUAVE Can Address Builder Centralization.”

1.2 Why Need Relay?

Relays act as intermediaries between block builders and validators in MEV-Boost. They are trusted by both parties, and their roles include the following:

• Preventing MEV stealing: If block builders were to directly provide block payloads to validators, there’s a risk that validators could create identical blocks themselves and steal the MEV. Relays prevent this by ensuring that validators only receive the block header and bid, without revealing the payload.

• Verifying block validity: Relays verify the validity of blocks before validators sign the block header. If validators sign invalid blocks, it could lead to missed slots.

• Ensuring data availability: After validators sign the block header, relays ensure that the block payload is delivered. Without relays, malicious builders could withhold payloads.

• Multiplexing: Relays allow block builders and validators to connect with a large number of counterparties. Without relays, these connections would have to be established manually.

1.3 Why Replace Relay?

1.3.1 Centralization Concerns

Although the situation has improved since the early days, a small number of relays still dominate block mediation in MEV-Boost. Currently, five relays—Flashbots, Ultra Sound (max profit), Ultra Sound (regulated), Bloxroute, and Gnosis—mediate around 90% of Ethereum blocks.

While these relays are operating fairly, what happens if a relay malfunctions or acts maliciously? The first concern is censorship. Relays have the ability to reject certain block builders' payloads. This isn't necessarily malicious but is already happening, as seen with Bloxroute (regulated), which rejects blocks containing illegal transactions in accordance with OFAC regulations.

The second concern is builder-relay collusion. Since relays have censorship power, they could choose to mediate blocks from only specific builders, regardless of bid size. This could increase the influence of certain builders, accelerating builder centralization.

The third issue is MEV stealing. Because relays can view builders' payloads, they could create identical blocks and steal the MEV themselves.

The fourth problem is liveness issues. If relays fail to release the block after validators sign the header, either maliciously or due to malfunction, it could result in an empty slot.

1.3.2 Latency Issues

Even if relays are not acting maliciously, their mere presence introduces latency into the block auction process. Latency arises during two phases: when the relay receives the block from the builder and when it runs simulations to verify the block's validity.

• Delivery latency: This is the time it takes for the relay to download the block from the builder, typically between 10 and 100 milliseconds. The closer the builder is to the relay geographically, the lower the latency, encouraging builders to colocate with relays.

• Simulation latency: This is the time it takes for the relay to simulate the block, averaging between 100 and 200 milliseconds.

Low latency is critical because, within the 12-second block auction window, if delivery latency is too long, there may not be enough time left to run the simulation, excluding the block from the auction.

Additionally, the computational cost of operating relays is quite high. While relays are manageable in the short term, in the long run, if Ethereum is to evolve into a truly decentralized computing platform and optimize block building efficiency, it would be ideal to either decentralize relays further or eliminate the need for them altogether.

2. Proposed Solutions

The Ethereum community has proposed several intriguing ideas to reduce the negative impact of relays or even eliminate the need for them altogether.

2.1 Optimistic Relays

Optimistic Relays aim to retain the presence of relays while significantly reducing latency. Similar to how optimistic rollups assume off-chain transactions are valid and process them accordingly, Optimistic Relays assume that the block provided by the builder is valid and proceed with the block auction. Three different versions of Optimistic Relays have been proposed.

Notably, the Ultra Sound relay has implemented Optimistic Relay v1, leading to increased processing of bids and reduced latency, which has boosted its market share.

2.1.1 v1: Asynchronous Block Validation

In Optimistic Relay v1, the relay does not immediately validate the block received from the builder but does so asynchronously. This significantly reduces simulation latency during the block auction process, effectively lowering overall latency. Additionally, it reduces the operational costs for relays, as they don't need to validate the surge of blocks at the auction's conclusion. To protect against malicious behavior, block builders participating in Optimistic Relays are required to stake a certain amount of funds, ensuring accountability if they attempt to submit invalid blocks.

2.1.2 v2: Header-Only Parsing

In Optimistic Relay v2, the relay receives only the bid details from the builder, rather than the entire block. This reduces the time required to download the block and eliminates the relay's ability to censor transactions since it no longer sees the payload. Similar to v1, builders are required to stake funds to deter malicious actions.

2.1.3 v3: Relay as an Oracle

Optimistic Relay v3 minimizes the relay's role, enabling direct peer-to-peer communication between builders and validators. Validators and builders each listen over the p2p network for header-only bids from builders and for validators' signed headers corresponding to those bids. In this version, the relay acts more like an oracle, detecting missed slots and imposing penalties when necessary.

Moving beyond v3, relays could eventually be replaced by a committee of validators, with these processes enshrined in the Ethereum network. This would bring the Ethereum network closer to achieving the final goal of ePBS (enshrined Proposer-Builder Separation).

2.2 MEV-Boost++ with EigenLayer

Although not primarily focused on relays, EigenLayer has proposed an interesting concept called MEV-Boost++ to address the issues caused by centralized relays. Unlike Optimistic Relays, which have been partially implemented, MEV-Boost++ remains a theoretical idea that has yet to be integrated into the block auction process.

The key innovation of MEV-Boost++ is that it allows validators to support partial block building by re-staking ETH through EigenLayer, in contrast to the current MEV-Boost system, which only supports full-block building.

In the current MEV-Boost system, validators only deal with full blocks because viewing builders' payloads before completing the block could lead to MEV-stealing. However, with MEV-Boost++, validators re-stake ETH through EigenLayer, which introduces slashing penalties if MEV-stealing occurs. This allows validators to inspect the payloads and complete block building by adding their own chosen transactions to the partial block.

MEV-Boost++ could solve two key problems caused by centralized relays:

• Preventing MEV-stealing: Validators face the risk of having their ETH slashed, making it difficult for them to collude with relays for MEV-stealing.

• Censorship Resistance: Even if a censoring relay refuses to process blocks containing certain transactions, validators can add these transactions to the partial block, bypassing the relay's censorship.

3. Enter TEE-Boost

Recently, a fascinating proposal was shared by Shea Ketsdever on the Flashbots forum, suggesting that relays could be completely eliminated by leveraging TEE.

3.1 What is TEE?

TEE, or Trusted Execution Environment, is a secure area within hardware like a CPU, isolated from the rest of the system, where sensitive data can be safely processed. TEEs ensure that trusted code can run securely, even if the external environment is compromised. Notable implementations include ARM’s TrustZone and Intel SGX. An example from daily life is the processing of biometric data, such as fingerprints or facial recognition, on mobile devices within a TEE.

TEE is designed so that even the operating system or programs with administrator authority cannot access the secure area. To verify that the code running inside the TEE is trustworthy, an attestation is used. Through the attestation process, the TEE can be verified as being in a trustworthy state, free from tampering or attacks. In the Intel SGX attestation process, a hash value representing the code and data within the SGX is computed. Additionally, a private key is generated and managed by the hardware, making it easy to prove the integrity of the code.

3.2 TEE + MEV-Boost

Now, let’s explore how combining TEE and MEV-Boost could remove relays.

Builders participating in TEE-Boost would begin by submitting an attestation at the start of each epoch, which validators would then verify. This attestation confirms that the builders are using verified software capable of producing valid blocks.

Subsequently, when builders submit blocks, they only need to generate a much simpler proof. Since builders have already been verified through their participation in TEE-Boost as using valid block-producing software, they need only submit a proof signed with a private key associated with the TEE. Validators can then confirm the validity of the block without needing to fully re-validate the attestation.

Because the TEE proof guarantees the block’s validity, builders don’t need to reveal the block to anyone, including relays. This approach effectively replaces the relay’s roles of preventing MEV-stealing and ensuring block validation. However, TEE-Boost does not entirely replace all relay functions. It doesn’t handle multiplexing (efficiently connecting builders and validators) or prevent malicious builders from withholding blocks. To address these issues, the Flashbots team has suggested some solutions and is seeking feedback from the community.

4. Final Thoughts

MEV-Boost has had a significant impact on the Ethereum ecosystem, with around 90% of the network’s blocks now being generated through it. While MEV-Boost is crucial to the roadmap of Proposer-Builder Separation, the ultimate goal should be to eliminate centralized relays. Similar to how Ethereum scaling solutions include both optimistic and validity rollups, this article explored two approaches for verifying the validity of block builders' payloads: Optimistic Relays and TEE-Boost. Although TEE may not be a perfect solution in terms of integrity, it could serve as a viable interim measure on the path to achieving full PBS.

We are already in the midst of the fourth major market cycle (perhaps nearing its end). The crypto market has grown significantly, with BTC ranking 9th ($1.26T) and ETH 25th ($409B) in overall asset rankings.

This naturally raises the question: Has a market of this size truly found Product-Market Fit (PMF)? Back in 2020-2021, most would have said no. However, given the market's maturity and the emergence of various protocols, many might now say yes.

My answer is mixed. There are indeed protocols that generate significant revenue even when accounting for token incentive spending, which makes me inclined to say yes. However, I must point out that the PMF of most of these protocols heavily relies on speculation. In contrast, protocols with little connection to speculation often fail to find widespread PMF, serving only a niche group of users.

Recently, many, including Vitalik Buterin, have raised similar points on social media. Even protocols that seem to have found PMF, mainly infrastructure ones, often their PMFs come from speculation. During the third bull cycle, there were many blueprints for using blockchain technology to solve real-world problems, with buzzwords like metaverse, P2E, and decentralized social networks capturing people's attention. However, despite the market's growth, it now appears that the vision of blockchain is shrinking, with only a few remaining degens and no real-world problems being addressed.

1. Was It All About Speculation?

It is natural for speculation to arise in an emerging industry. While speculation can lead to many victims, it also helps the market and industry achieve scale. In other words, for speculation to be justified, the industry must ultimately find a suitable PMF.

Throughout this bull cycle, it seems the industry's efforts to find PMF have regressed. Despite the influx of talent and capital leading to significant advancements in regulation, technology, and infrastructure, there are still no blockchain products with widespread PMF. Even with the approval of Bitcoin and Ethereum ETFs, discussions around visionary concepts like decentralization and metaverse have dwindled since the 2021 bull ycle, and the market appears to be targeting increasingly niche segments.

Was the market's growth ultimately driven solely by speculation? To find the answer, I divided the market into three periods.

2. Answer: Mostly

2.1 Internet Money

After the concepts of Bitcoin and blockchain first emerged in 2008, Bitcoin was primarily used as a payment method for online transactions due to its censorship resistance and ease of cross-border payments.

A notable example is the use of Bitcoin in MMORPG games with active economies, like World of Warcraft, for trading items. Additionally, Bitcoin was used for illegal transactions on dark web marketplaces such as Silk Road, involving drugs, weapons, and pornography.

Despite its significant use in illegal transactions, Bitcoin found its PMF with specific groups even when it was not widely known.

2.2 Speculation

During this period, cryptocurrencies were largely viewed as speculative assets. Although projects like Steemit, Livepeer, Filecoin, and Brave Browser aimed to solve real-world problems, the market was heavily speculative.

At the end of 2013, Bitcoin's price surged from $100 to $1100, reinforcing its perception as a speculative asset. This led to numerous Ponzi schemes like OneCoin and many victims.

The first bull cycle in 2013 failed to attract widespread attention, but the second bull cycle in 2017 captured global interest. BTC and ETH achieved significant market capitalizations, with speculative trading especially prevalent in the South Korean market. During this period, projects like EOS, ADA, TRX, and BNB raised substantial funds through ICOs, though many ICO scams emerged from projects with no real substance.

Since the market was built on speculation, the subsequent crash led to a prolonged crypto winter. However, projects built during this time and quantitative easing post-COVID-19 helped the market recover in 2021. DeFi protocols like Uniswap and Compound boomed on-chain, with speculation active both off-chain and on-chain.

This period saw heightened interest in blockchain technology itself, with many idealistic projects aiming to solve problems through decentralization. While grand visions like the metaverse, P2E, and decentralized social mostly remained unrealized, they inspired many people.

2.3 Speculation Infrastructure

Following the 2021 third bull cycle, the crypto industry attracted massive interest, leading to increased efforts to incorporate blockchain into traditional Web2 industries to find PMF. Within the Web3 scene, venture capital investments grew, and more teams started building projects that addressed real-world problems beyond speculation. These teams focused on improving scalability, interoperability, and UI/UX to achieve mass adoption of blockchain technology.

These efforts tackled crucial issues. Notable advancements include bridges (e.g., Across, Wormhole, LayerZero) addressing fragmented liquidity, and layer 2 solutions (e.g., Optimism, Arbitrum, Polygon) effectively solving base layer scalability problems.

Some of these protocols generate more fee revenue than they spend on token incentives. A representative example is Base. Layer 2 business models rely on providing highly scalable blockspace that depends on Ethereum's security. They pay gas fees for storing data on the Ethereum network and charge users transaction fees. Without governance token incentives, Base achieved an impressive gross profit of $35M over the past 180 days.

Additionally, numerous projects within the on-chain ecosystem provide utility to users, with the following protocols achieving a degree of PMF:

• L1: Ethereum, Solana, Tron

• L2: Arbitrum, Base, Optimism

• Bridge: LayerZero, Wormhole

• Staking: Lido, Rocket Pool, Jito

• Restaking, LRT: EigenLayer, etherfi, Symbiotic

• DeFi: Aave, Maker, Uniswap, Pendle, Ethena

• NFT: OpenSea, Zora

• Prediction Market: Polymarket, Azuro

• Social: Farcaster, ENS

• Infrastructure: Chainlink, The Graph

• Meme: Pump Fun, Moonshot

Here’s my hot take

While the protocols mentioned above do provide significant utility to users and have achieved PMF, I believe that, at this point, much of this PMF ultimately revolves around speculation. Alternatively, services that have achieved PMF with little connection to speculation tend to have a very limited audience.

• The essence of smart contract L1 is to conduct computations in a decentralized environment, providing benefits like censorship resistance and liveness. However, genuine use cases aligning with this essence are rare, with most users utilizing L1 as platforms for speculation.

• The core purpose of L2 is to offer fast scalability while relying on the security of the base layer. Although L2 has certainly achieved PMF, much of the demand stems from users wanting to speculate on-chain more quickly and cheaply. If L1s are high-stakes, expensive casinos, L2s are their lower-stakes, more affordable counterparts.

• Bridges facilitate the movement of capital and messaging between various networks, making them essential infrastructure in the current landscape of numerous networks. Without bridges, many users and businesses would face significant inconvenience. However, like L2s, bridges are often used by users seeking speculative opportunities on different networks, like services that transfer funds between casinos.

• Staking and restaking are critical for the security of protocols and have been highly successful in terms of TVL. While seeking incentives is natural and not inherently wrong, many investors participate with expectations of unsustainably high rewards (airdrops, yield, etc.).

• DeFi enables anyone to engage in financial activities on-chain. Although there are increasing cases of integrating RWA, the market remains small, and many DeFi protocols are linked to speculation. For instance, Pendle and Ethena have grown rapidly by finding appropriate PMF, but this growth has been driven by user speculation. Both protocols attracted significant user and TVL by leveraging expectations of airdrops.

• The NFT marketplace sector vividly illustrates the impact of speculation. NFT marketplaces are neutral platforms for trading NFTs, but examples like OpenSea and Blur show a sharp decline in trading volume once the NFT speculation frenzy wanes or token incentive programs end.

• Web3 social aims to address the issues of centralized social media. While there is some user expectation of speculation, this sector is one of the few where the building intent and actual PMF align. However, it remains a niche market as not many people yet share the concerns about centralization of Web2 social.

• On-chain infrastructure like oracles and query services are essential for the safe and efficient operation of the on-chain ecosystem, but they are still largely used in services related to speculation.

• Prediction market and meme-related protocols inherently aim to facilitate speculation.

For example, imagine you buy YT-eETH through Pendle on the Arbitrum network. Arbitrum, a Layer 2 solution, reduces your costs and time. Pendle allows you to separate the yield and principal of eETH, offering various strategies. Etherfi restakes and mints liquid ETH on your behalf, and EigenLayer lets you stake ETH across multiple protocols simultaneously. While these services are useful, their activities are driven by the speculation surrounding AVS rewards and potential airdrops.

p.s. There are blockchain-related services widely used in real life, but they typically follow Web2 paradigms, with blockchain as just one feature among many. Examples include Reddit's avatar NFTs and Sweatcoin.

Don’t get me wrong

In a free market, products don’t necessarily have to be used as intended. Even if a product generates demand and revenue through avenues like speculation, it is still valuable. However, if the PMF doesn’t align with the core essence of blockchain, then blockchain might not be necessary. Traditional Web2 technology can often suffice.

Given the market's size, why haven't we seen widespread PMF in blockchain products? It's because modern society doesn't truly need blockchain yet.

3. From Speculation to Credible Neutrality

As Josh Stark explains in Atoms, Institutions, Blockchains, the value blockchain brings to the world is credible neutrality in the digital realm, akin to the roles of physical laws and social norms. Physical laws are essential as they define space, time, and matter. Similarly, social norms, such as governments and laws, are necessary as they define interactions within human society. Conversely, modern society doesn't yet need blockchain as digital interactions are still relatively well-coordinated based on trust in centralized entities.

There are exceptions, though. In countries where social norms fail due to government corruption or outdated infrastructure, Bitcoin and stablecoins can play crucial roles in the economy. This is evident in Latin America and Africa. Unlike people in developed countries who view crypto as an investment, residents in these regions use crypto to sustain their livelihoods. Here, blockchain's credible neutrality gives Bitcoin and stablecoins the characteristics of assets and money, finding real PMF beyond speculation.

To find broader PMF based on credible neutrality, we can only wait for more failures of centralized systems. Although not directly related to blockchain, Trump’s Truth Social emerged from a desire to avoid Big Tech censorship. While such centralized system failures are unfavorable for developed nations, they could eventually drive people toward blockchain systems. Essentially, blockchain technology will provide true utility beyond speculation when the flaws of centralized systems become evident.

However, issues like social media censorship, data breaches, and cloud service outages are insufficient catalysts. While these problems occur, the benefits of centralized services still outweigh them, leading most people to continue using existing systems. As I mentioned in a previous article, the biggest catalysts for blockchain finding PMF based on credible neutrality will be 1) the failure of the dollar and 2) rapid advancements in AI. Recent endorsements of Bitcoin by prominent figures like Trump, Larry Fink, and Jamie Dimon reflect a similar trend.

4. Final Thoughts

Over the past three years, blockchain technology and the industry have grown rapidly. This growth has been primarily driven by investor speculation. While speculation often gets a bad rap, we should recognize its contributions to the industry’s development. However, it's disappointing that the current PMF in the blockchain market remains heavily focused on speculation. There’s little we can do to find fundamental PMF based on credible neutrality.

Despite this, I remain very optimistic about the blockchain industry. As Balaji pointed out, the world is in a continuous cycle of bundling and unbundling. As our social systems become increasingly centralized, they are bound to encounter issues, and the demand for unbundling will grow. I hope that, in the future, blockchain will play a crucial role in protecting human sovereignty.

What is the purpose of blockchain technology? While it can be utilized across various fields such as social networks, gaming, and commerce, I believe, as discussed in my previous article “Will it be the same as ever?: Money, AI, and Blockchain,” that the ultimate endgame for blockchain is to serve as a hard currency and an anti-thesis to AGI.

Building on this idea, numerous projects have emerged aiming to address the issues within the centralized AI industry through the decentralized nature of blockchain. These projects are typically referred to as decentralized AI (dAI) initiatives. A major development in the dAI sector occurred recently when Sentient, a project with Sandeep Nailwal, co-founder of Polygon, as a core contributor, raised $85 million in seed funding.

As the AI industry advances, leading language models like GPT-4 and Claude 3.5 Sonnet are mostly operated as closed-source. Sentient aims to counter this by building community-driven open-source AI models. By leveraging blockchain protocols, developers can 1) monetize their models, 2) collaborate to collectively build AI models, and 3) become stakeholders in an Open AGI economy. According to Sandeep, Sentient will be built on the Polygon AggLayer, suggesting it will be a zk L2 based on Polygon CDK.

During my research on Sentient, I came across an intriguing project called SAKSHI. Clicking the research tab on Sentient’s website automatically redirects to the Open AGI Research forum, where the post “SAKSHI: Decentralized AI Platforms” is featured. This post caught my attention primarily due to its authors, which include Sreeram Kannan (founder of EigenLayer), co-founders of Sentient, and professors from prestigious universities.

1. SAKSHI: Decentralized AI Platforms

1.1 EigenLayer, Babylon, … SAKSHI?

Recently, there has been an increase in prominent university professors and students from the United States building projects in the crypto scene. Historically, examples include Silvio Micali from MIT, UPenn, UToronto, and Tsinghua University with Algorand, and Emin Gun Sirer from Cornell University with Avalanche. More recently, notable examples are Sreeram Kannan (University of Washington) with EigenLayer and David Tse (Stanford) with Babylon.

In my opinion, compared to the AI industry, the blockchain industry lacks significant academic involvement, which implies that examples above are very promising. So, what are the professors involved in Sentient and SAKSHI aiming to build? Before diving into the projects, let’s take a look at the backgrounds and research areas of the professors involved.

Sentient

• Pramod Viswanath: Forrest G. Hamrick Professor in Engineering in the Department of Electrical and Computer Engineering, Princeton University (Blockchain, Deep Learning, Wireless Communication)

• Himanshu Tyagi: Associate Professor in the Department of Electrical Communication Engineering, Indian Institute of Science (Blockchain, Privacy, Federated Learning, Statistics, …)

• Sewoong Oh: Professor, Allen School of Computer Science & Engineering, University of Washington (Machine Learning, Federated Learning)

SAKSHI

• Suma Bhat: Assistant Professor in Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign (Machine Learning, Natural Language Processing)

• Zhixuan Fang: Assistant Professor in the Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University (Blockchain, Collaborative Learning, Network Economics)

• Sreeram Kannan: Affiliate Associate Professor in the Department of Electrical & Computer Engineering, University of Washington (Blockchain)

• Xuechao Wang: Assistant Professor in Thrust of Fintech at HKUST (Blockchain, DeFi)

• Pramod Viswanath: Forrest G. Hamrick Professor in Engineering in the Department of Electrical and Computer Engineering, Princeton University (Blockchain, Deep Learning, Wireless Communication)

• Himanshu Tyagi: Associate Professor in the Department of Electrical Communication Engineering, Indian Institute of Science (Blockchain, Privacy, Federated Learning, Statistics, …)

All are renowned university professors, and experts in either blockchain or AI. Given the recent success of notable professor-led projects like EigenLayer and Babylon, I couldn't help but be intrigued by SAKSHI.

1.2 Motivation & Aim of SAKSHI

The issue the professors aim to solve is straightforward: the centralization of AI. In the future, most computing is expected to be used for inference rather than training, and currently, inference is highly centralized. Companies like OpenAI and Anthropic offer closed-source LLMs through web interfaces or APIs, raising potential issues of privacy, transparency, and rent-seeking.

SAKSHI aims to create a decentralized marketplace for inference using blockchain. Clients needing inference can request tasks on SAKSHI, and numerous AI suppliers can provide AI models and computing power on the platform, earning appropriate rewards. SAKSHI addresses several potential issues with this open marketplace approach:

1. Insufficient Clients for Individual Suppliers: SAKSHI introduces aggregators to collectively provide services on behalf of suppliers.

2. Poor Quality or Irrelevant AI Models: SAKSHI uses proof of inference to verify that tasks are completed correctly.

3. Non-Payment by Clients: SAKSHI enforces payments through SLA contracts and proof of service delivery via smart contracts.

In essence, SAKSHI’s core goal is to establish an open aggregate marketplace for inference, where clients and suppliers can freely transact, enhancing decentralization and transparency by removing the need for third-party trust through blockchain, smart contracts, and proof systems.

1.3 The Six Layer Architecture

SAKSHI consists of six layers: Service Layer, Control Layer, Transaction Layer, Proof Layer, Economic Layer, and Marketplace. The first two layers are web2 components, while the remaining four are related to blockchain. A brief explanation of each layer is provided below, with detailed descriptions covered in “2. Architecture of SAKSHI”:

1. Service Layer: Facilitates the exchange of inference services between clients and servers.

2. Control Layer: Matches clients with servers based on server network state and client requests.

3. Transaction Layer: Handles payments for services.

4. Proof Layer: Resolves disputes arising from inaccurate inferences or AI model duplications.

5. Economic Layer: Ensures the economic security of the SAKSHI platform.

6. Marketplace: A decentralized platform for buying and selling inference services.

1.4 Process in a Nutshell

Before delving into detailed explanations of each layer, let's take a high-level look at how the protocol operates. Given that SAKSHI includes an Aggregator to mediate between Clients and Servers, the first step involves signing SLAs (Service Level Agreements) through smart contracts between Clients and Aggregators, and Aggregators and Servers (Transaction Layer).

When a Client requires AI inference, they send a request via the API (Service Layer). The Aggregator then matches this request to an appropriate Server based on the AI models, computing power, and other relevant factors available (Control Layer). The Server provides the AI model and inference service, receiving appropriate compensation. If the inference is inaccurate, challengers can raise disputes, which are handled within the Proof Layer.

2. Architecture of SAKSHI

2.1 Service Layer

The Service Layer facilitates the actual exchange of inference services between clients and servers (AI suppliers), similar to a traditional web2 server-client architecture. When a client sends an inference query, the Control Layer matches it with the appropriate server. Once the server provides the AI model and inference service, payments are transferred from the client to the server via the Transaction Layer. To prevent malicious activities, signed inference requests, output data posted on the decentralized application (DA) layer, and previously exchanged micropayments can be used for dispute resolution.

2.2 Control Layer

The Control Layer tracks the network state, including model capacity, hardware capacity, request load, and location of servers. It also monitors SLA contract information from the Transaction Layer between client-aggregator and aggregator-server, ensuring appropriate client-server matching based on this data.

2.3 Transaction Layer

The Transaction Layer handles payments for inference services. Since SAKSHI is a decentralized system, it ensures seamless payments through smart contract-coded Service Level Agreements (SLAs), facilitating payments upon service delivery. This layer utilizes decentralized middleware provided by Witness Chain.

2.4 Proof Layer

The Proof Layer resolves disputes related to malicious activities within SAKSHI. It includes Proof of Inference, which verifies the accuracy and validity of AI model computations, and Proof of Model-ownership, which addresses intellectual property disputes. Witness Chain AVS operators act as challengers in these disputes.

Proof of Inference

There are two methods for verifying the accuracy of inference: zkML (zero-knowledge machine learning) and opML (optimistic machine learning). zkML is impractical due to its time and cost inefficiency for complex AI computations.

SAKSHI adopts opML, where challengers contest the validity of computations by re-running models. To manage large computations, SAKSHI employs Arbitrum's bisection scheme, which breaks down large computations into smaller parts.

AI models, often non-sequential, are structured as Directed Acyclic Graphs (DAGs) in SAKSHI to identify faulty entries efficiently.

Proof of Model-ownership

To address the issue of malicious actors copying and monetizing open-source AI models, SAKSHI requires AI model suppliers to embed watermarks during training. These watermarks, committed to the blockchain, help trusted judges verify if an AI model has been stolen. This system, however, is limited to the SAKSHI platform and cannot prevent external misuse of open-source models.

2.5 Economic Layer

In SAKSHI, incentivized challengers report malicious activities, ensuring fair behavior among entities in a decentralized system. Instead of issuing its governance token, SAKSHI leverages EigenLayer, which utilizes the crypto-economic security of staked ETH, providing substantial security. Protocols using EigenLayer are known as AVS, with Witness Chain serving as an AVS for the Transaction Layer. The security of SAKSHI is expected to depend on the re-staked ETH delegated to Witness Chain.

3. Final Thoughts: Computing as the Future Currency

Kojo Osei, a partner at Matrix VC, recently predicted that the cost of inference will approach zero due to increasing competition among AI models and the enhanced performance of consumer device hardware. I agree, anticipating that in a few years, decent AI model inference will be feasible on user devices.

If this prediction holds true, computing could become a future currency, as suggested by Sam Altman. Similar to torrent services, individuals could earn rewards by contributing idle computing power from their devices to the network. For this to happen, a decentralized open marketplace must exist, allowing transparent participation and rewards. Will SAKSHI play a pivotal role in this future?

1. The Next Challenge for Ethereum: Builder Centralization

Ethereum is often regarded as one of the most decentralized networks alongside Bitcoin. Due to the relatively low hardware requirements for operating an Ethereum node, almost anyone can run a node. There’s some redundancy though, the network boasts over 1 million validators.

However, a critical issue often overlooked is builder centralization. Builders are the entities that gather transactions and bundles to create blocks on the Ethereum network. Over the past seven days, 95% of blocks were generated by just three builders.

Despite this, as Vitalik Buterin has pointed out, builder centralization does not pose a severe threat to the overall security of the Ethereum network. This is because, even if block building is somewhat centralized, the validators(proposers) who verify these blocks remain decentralized. Nonetheless, builder centralization can lead to various problems such as censorship, rent-seeking, and liveness issues.

This article will explore the journey of Flashbots in addressing the negative externalities of Ethereum's MEV and examine how SUAVE could ultimately resolve issues related to MEV, including builder centralization.

2. Progress So Far

2.1 Proof of Work

Before The Merge upgrade, the Ethereum network operated on PoW consensus, similar to the Bitcoin network, where miners used hardware to mine blocks. During this period, when searchers identified MEV opportunities in the mempool, the only way to get their transactions or bundles included in a block was through a priority gas auction (PGA), where they bid higher gas fees than other searchers.

There were fundamental problems with this approach. First, MEV stealing was an issue. Miners could see the contents of the transactions or bundles submitted by searchers and, instead of including them in the block for a priority fee, they could copy these transactions and steal the MEV themselves. Thus, searchers had to trust miners to earn MEV profits.

The second problem was network congestion. Whenever MEV opportunities arose, searchers competed by bidding higher priority fees, which led to increased congestion on the Ethereum network. This made average transaction fees expensive and unpredictable, negatively impacting regular users.

To address the negative externalities of MEV on the PoW Ethereum network, Flashbots introduced the Flashbots Auction, consisting of mev-geth and mev-relay. The key components were: 1) whitelisting miners, 2) establishing a private mempool, and 3) implementing a sealed bid auction system.

Users and searchers could submit transactions or bundles to the private mempool of Flashbots Auction, which were then sent to whitelisted miners using the mev-geth client via a centralized mev-relay. Searchers expressed bids for their bundles, and miners used mev-geth to include the highest bidding bundles in the block.

Unlike the previous system, searchers used a private mempool, so their actions didn't impact the Ethereum gas market, and they couldn't see other searchers' bids, reducing competition. Consequently, Flashbots Auction effectively reduced congestion on the Ethereum network. However, whitelisting miners was still necessary because they could still see the content of bundles submitted by searchers.

Flashbots Auction became widely adopted, with over 90% adoption of mev-geth. This significantly reduced failed MEV transactions and lowered average gas fees on the Ethereum network, effectively mitigating many of the negative externalities associated with MEV.

2.2 Proof of Stake (PoS)

In September 2022, the Ethereum network transitioned from PoW to PoS with the activation of The Merge upgrade. The process of including user-submitted transactions in blocks remained largely unchanged from PoW. However, there was a critical issue with adopting Flashbots Auction directly: whitelisting.

In PoW Ethereum, miners physically owned their hardware, making the whitelisting process relatively straightforward. However, with the switch to PoS, a wide range of entities could participate in validating anonymously, making whitelisting extremely difficult.

To address the negative externalities of MEV on PoS Ethereum, Flashbots introduced a new protocol called MEV-Boost. The Ethereum network roadmap includes the PBS (Proposer-Builder Separation) upgrade to decentralize MEV, and MEV-Boost implements part of PBS.

In this new setup, block builders receive transactions and bundles from users and searchers to create the most valuable full block, while proposers select the highest bid full block from block builders and propagate it to the network. Unlike mev-geth, MEV-Boost acts as a sidecar to the consensus client, making it compatible with any client type.

Here’s how MEV-Boost operates:

1. Block builders receive transactions from searchers and private order flow, using their MEV extraction algorithms to reorder transactions for maximum profitability. They then create a full block and submit a bid to the relay.

2. The relay verifies the validity of blocks received from builders and stores them.

3. The relay sends the block headers, along with bids, to the proposer.

4. The proposer selects the block header with the highest bid from those sent by the relay and signs it.

5. The relay reveals the full block content corresponding to the signed header to the proposer.

6. The proposer submits the complete block to the network and collects the bid attached by the block builder.

From the perspective of Ethereum validators, MEV-Boost offers a significant advantage: there is no need for a whitelisting process. Validators simply run Flashbots' MEV-Boost, and block builders just extract the highest value MEV and submit it as bids. This means validators can earn MEV revenue without needing their own MEV extraction algorithms. Consequently, MEV profits are decentralized rather than being concentrated among a few entities.

Despite being an external middleware rather than a built-in protocol, MEV-Boost has been successfully adopted by over 90% of Ethereum validators for an extended period. While there is a drawback that builders and proposers must trust the relay, the number of relays has increased to eight, reducing Flashbots relay's dominance and alleviating related concerns such as censorship.

3. Builder Centralization

3.1 Why Builders Tend Toward Centralization

While MEV-Boost has mitigated many of the negative externalities associated with MEV, the issue of builder centralization, mentioned earlier, remains unresolved. Currently, around 90% of Ethereum network blocks are created by just three to four block builders. But why do block builders tend to centralize? There are two main reasons:

Exclusive Order Flow (EOF)

Firstly, the block builder market is fundamentally a winner-takes-all market. Imagine you are a searcher who has identified an MEV extraction opportunity and bundled it. Which builders will you send your bundle to? While you could send it to all builders, the more builders you involve, the higher the risk of MEV stealing, as builders can see the content of the bundle. Therefore, your optimal strategy would be to send the bundle only to the top few builders with the highest probability of block inclusion.

The graph above shows that builders receiving more bundles from searchers have a higher probability of block inclusion. This phenomenon accelerates the centralization flywheel: if a builder receives more bundles from searchers, it is more likely to build more profitable blocks. Consequently, these blocks are more likely to be adopted by proposers on the Ethereum network, incentivizing more searchers to send their bundles to that builder. Sending bundles to less dominant builders could result in delays in block inclusion, making gas fee predictions difficult and potentially losing MEV extraction opportunities.

Beyond this natural tendency towards centralization, builders can source additional transactions or bundles through EOF. For example, a specific builder might offer privacy guarantees or a share of the extracted MEV to users and searchers who send transactions or bundles exclusively to them. This additional order flow, inaccessible to other builders, further accelerates builder centralization.

Indeed, as shown in the graph, BloXroute has a significantly higher block inclusion rate compared to its peers. This is because BloXroute operates not only as a block builder but also as a relay service, giving it a latency advantage in processing transactions. Additionally, BloXroute sources EOF through services like BackRunMe.

BackRunMe allows users to submit private transactions, protecting them from malicious attacks like front-running and sandwich attacks. Moreover, if MEV profits are generated from backrunning the private transactions submitted to BackRunMe, the profits are distributed according to the ratios shown in the chart. Users and searchers can enjoy various benefits by using BackRunMe's swap UI or simply changing their RPC to submit transactions.

So, what can new block builders do? Unfortunately, they have limited options other than increasing their market share at a loss or offering services to attract users and searchers' EOF. The former approach, known as a block subsidization strategy, involves setting higher bids than the MEV profits generated from building blocks to increase the block inclusion rate. For example, the f1b builder successfully used this strategy to quickly increase their searcher count.

Cross-Domain MEV

The more order flow a block builder has access to, the higher the probability of generating more profitable blocks. If some block builders also create blocks for other networks, they can access not only the order flow from the Ethereum network but also external order flow. This capability would likely lead to further centralization around these builders.

3.2 What Should We Do?

We've explored why the builder market tends to centralize. Although centralization of builders does not pose a severe security threat due to the decentralized nature of proposers (validators) who verify and propagate blocks, it can still lead to issues such as 1) censorship, 2) rent-seeking, and 3) liveness problems.

Censorship could potentially be addressed by future Ethereum protocol features like crList, which would natively force builders to include all transactions as required by proposers. However, addressing rent-seeking in a monopolistic market and resolving liveness issues due to downtime is more challenging.

Therefore, the best solution is to prevent builder centralization in the first place by mitigating its main causes—EOF and cross-domain MEV. To address these issues, Flashbots introduced the Single Unifying Auction for Value Expression (SUAVE) protocol. (It's worth noting that SUAVE is not the only potential solution to builder centralization; for a variety of other potential solutions, see Jon Charbonneau’s 'Decentralizing the Builder Role').

4. Here Comes SUAVE

4.1 TL;DR

SUAVE focuses on addressing the two main factors contributing to builder centralization: EOF and cross-domain MEV. Firstly, SUAVE can accept transactions from all networks, enabling decentralized builders to inherently extract cross-domain MEV. Secondly, SUAVE optimizes conditions for users by privately handling preferences and offering a share of MEV profits.

4.2 Overview

SUAVE is a separate blockchain from the Ethereum network, offering a plug-and-play mempool and decentralized builder service that can be used by multiple networks. This allows other networks to outsource the complex processes of mempool management and decentralized block building to SUAVE. SUAVE consists of three main components:

Universal Preference Environment

Users and searchers submit transactions, bundles, intents, and other expressions of preferences to SUAVE's mempool, instead of the original network's mempool, along with their bids. In SUAVE, these preferences are treated as a native transaction type. By aggregating preferences from various domains into a single mempool, the probability of optimal execution increases. This setup benefits builders by lowering entry barriers and increasing potential profits.

Optimal Execution Market

Executors (Searchers, Builders, etc.) monitor the SUAVE mempool and compete to create bundles with the best execution conditions. A key concept introduced here is the order flow auction (OFA).

In the traditional MEV-Boost model, MEV profits flow in a single direction from users to searchers to builders to proposers. However, with OFA, executors compete for users' preferences, allowing users to also receive a share of the MEV profits. This strategy is similar to services like BackRunMe, which aim to attract more EOF by redistributing some MEV profits to users and searchers. Additionally, SUAVE ensures the privacy of preferences in its mempool, protecting them from malicious MEV attacks.

The difference is that, while such strategies can lead to the centralization of specific builders in the current builder market, SUAVE embeds OFA into the protocol itself, giving all decentralized builders access to these preferences. The concept of OFA, as proposed by Flashbots, is already implemented in the Ethereum network through MEV-Share and will later be incorporated into SUAVE.

Decentralized Block Building

In the previous components, most preferences find their optimal execution route. Decentralized block builders then use this information to construct partial or full blocks that maximize MEV profits, which they then pass on to validators of various networks.

Not all validators of other networks may use SUAVE, similar to how not all Ethereum validators use MEV-Boost. Validators listening to SUAVE can accept SUAVE blocks and add profitable blocks to their network. If they are SUAVE-unaware, SUAVE's block builders must participate in a priority gas auction (PGA) to get their blocks included. Once preferences are fulfilled in the destination chain, an oracle notifies the SUAVE network, and the bid is sent to executors for settlement.

4.3 MEVM

SUAVE is a blockchain that uses MEVM as its execution environment. The MEVM is built on the EVM framework, with added precompiles for MEV use cases. Developers can use Solidity to create MEV applications as smart contracts, enabling the decentralized building of previously centralized MEV-related infrastructure. For example, different methods of block building or order flow auctions can be implemented as smart contracts.

Given the need for sensitive data and computations, MEVM also offers privacy features. Sensitive computations are executed off-chain by execution nodes. Initially, Flashbots or third parties will provide this in centralized way but eventually, it will be executed in trusted execution environments (TEE) like Intel SGX.

5. Summary & Challenges Ahead

In summary, SUAVE aims to collect transactions from all blockchain networks and provide blocks with the most efficient execution to those networks. If SUAVE's vision is fully realized, it will enable true decentralization of MEV, offering the following benefits to various participants in the blockchain ecosystem:

• Users: Protected from malicious MEV attacks through privacy and offered the best execution.

• Builders: Can compete fairly with other builders due to SUAVE’s inherent privacy transactions and order flow auctions (OFA), and have access to cross-domain preferences, enabling them to build more profitable blocks than when operating within a single domain.

• Networks: Can easily outsource the block-building process to SUAVE.

Despite its ambitious vision, SUAVE is still in its early stages and faces several challenges before it can be fully realized.

• Security Model: SUAVE’s security model is still undefined. Given that SUAVE blocks are likely to be used in Ethereum and Ethereum-based L2 networks, its security level ideally should match that of Ethereum, but achieving this is complex. There are discussions on whether SUAVE should be built as an Ethereum L2 or use EigenLayer’s crypto-economic security.

• Atomic Cross-Domain Transactions: These are not guaranteed. It is challenging to process transactions atomically across networks with different block times. A transaction might succeed on a fast-block-time network but fail on a slower one. Additionally, since not all validators on all networks are SUAVE-aware, including blocks via a priority gas auction (PGA) might fail.

• Oracle Design: A sophisticated oracle design is needed to accurately and swiftly bring results from external domains into SUAVE for settlement. Oracles must be at least as secure as SUAVE, as they can become attack vectors.

• User Experience: A user-friendly UX must be designed for SUAVE. Users need to set bids for their preferences and hold ETH in the SUAVE network. An interface that allows users to easily express various types of preferences is also necessary.

The biggest concern is whether SUAVE can achieve a significant adoption rate similar to mev-geth or MEV-Boost. For SUAVE to realize its vision, it must achieve economies of scale. Many users from numerous networks need to send their preferences to SUAVE, and numerous builders must participate to create an efficient system. While mev-geth was a client and MEV-Boost was a middleware sidecar that existing validators could easily adopt, SUAVE is a blockchain network based on MEVM. Therefore, it remains to be seen whether this large system can achieve meaningful adoption across many networks.

Human beings are remarkable creatures. While the pace of biological evolution is incredibly slow, the rate at which humans transform the world through science and technology is astonishingly rapid in comparison. Consider the contrast between our lives today and those of people a thousand years ago. Despite having similar appearances and not vastly different cognitive frameworks, the disparity in living standards is immense.

However, no matter how swiftly the world changes, humans are ultimately bound by their physical and genetic makeup, composed of organic and inorganic materials. The instinct-driven struggles for wealth and power, class conflicts, wars to re-establish international order, and cycles of wealth and debt have been persistent throughout history and will likely continue. The ways humans react and behave in response to these issues are unlikely to change significantly over time.

This perspective suggests that by examining historical human actions and responses to major events, we can anticipate future patterns. While we cannot predict the future with absolute certainty, unless there are dramatic changes in human biology or a radical shift in our collective mindset, such as a universal conversion to Buddhism achieving enlightenment, we can use the past to make educated guesses about future trends.

Numerous books have been published analyzing the unchanging aspects of human society and our consistent reactions to historical events. For instance, Morgan Housel's "Same as Ever" offers insightful explanations about the persistent nature of human thought processes from a micro perspective. On the other hand, Ray Dalio's "Principles for Dealing with the Changing World Order" provides a macro perspective, analyzing the repetitive history of empires. Both books are highly recommended for readers interested in understanding these enduring patterns.

In this context, this essay aims to explore the significant, unavoidable trends humanity currently faces and their potential impacts on society, drawing parallels with historical precedents. Among these trends, I focused on the wavering status of the US dollar and the rise of Artificial General Intelligence (AGI), noting their commonality in presenting significant risks due to centralization. Consequently, I believe that blockchain technology, which inherently promotes decentralization, will play a crucial role in the future of human society. Each section of this essay will delve into how the blockchain industry, led by Bitcoin, might ultimately shape our world.

1. Same as Ever: Money

1.1 Reserve Currency Collapse is Inevitable

Currency is a social contract established for the purpose of facilitating barter. The legitimacy of this contract relies on social factors such as the balance of power within the international order and the trust of its participants. Considering that there have been no significant changes in human thought and emotional systems over a long historical period, it is highly likely that future currency systems will follow historical precedents.

Most people living in the present day are already very familiar with the US dollar as the global reserve currency, using it in daily life without much question. The United States' dominance in military, financial, scientific, and various other fields has solidified the dollar's seemingly eternal status. However, humans have a tendency to be complacent about things they have not personally experienced. A brief exploration into the essence and history of money reveals that the tenure of a global reserve currency is often shorter than one might expect.

The US dollar has held its position as the sole global reserve currency only since the establishment of the Bretton Woods system in 1944, a span of merely about 80 years. Before assessing the current status of the dollar, it is instructive to briefly review the global reserve currencies that preceded it. Prior to the dollar, the British pound sterling served as the world's reserve currency, and before that, the Dutch guilder held this role.

The rise and fall of the Netherlands and Britain as great powers, and their tenure as holders of the global reserve currency, followed remarkably similar patterns. Both nations began their ascent by triumphing in wars against declining powers. This victory acted as a catalyst for their increasing national competitiveness, spurred by developments such as the growth of capitalism and the Industrial Revolution. These advancements laid the foundation for their status as reserve currency nations.

However, as history has repeatedly shown, the wealth and prosperity derived from holding the status of a global reserve currency often sow the seeds of decline. Increasing current account deficits and widening income inequality weaken national competitiveness and accelerate the accumulation of debt. Eventually, massive debts incurred through wars, along with the devaluation of their currencies, force these once-dominant nations to relinquish their reserve currency status to emerging powers.

The United States, currently the world's leading superpower, has followed a similar trajectory. After the Civil War, the nation enhanced its competitiveness through the Second Industrial Revolution, the development of capitalism, and its geopolitical advantages. Surpassing a declining Europe in wealth and prosperity during and after the First and Second World Wars, the U.S. reached new heights. As victory in World War II became certain, the United States convened a conference to restructure the post-war financial order, adopting the Bretton Woods system, which established the dollar as the reserve currency under the gold standard.

However, a reserve currency economy based on hard currency, like the gold standard, presents a dilemma. To use the dollar as the primary currency for international trade, there must be a sufficient supply of dollars, requiring the reserve currency nation to maintain a deficit. While gold reserves remained constant, the increasing issuance of dollars inevitably led to the currency's devaluation and eroded international trust in the reserve currency. This issue is known as the Triffin Dilemma.

The Cold War with the Soviet Union, the Vietnam War, and the Oil Shock exacerbated trade deficits and inflation. When the U.S. could no longer meet the demand for gold redemption, President Richard Nixon ended the gold convertibility of the dollar in 1971. This led to a dramatic rise in the price of gold from the fixed $35 per ounce to $850 per ounce by 1980, marking the beginning of the fiat currency era and an age of high inflation.

Fortunately, due to the unprecedented high-interest-rate policies implemented by Paul Volcker, which reached annual rates of 20%, and the successful establishment of the petrodollar system, the dollar regained its value. This recovery ushered in a period of economic prosperity for the United States during the 1990s.

However, the dynamics of dollar issuance underwent a complete transformation after the end of the Bretton Woods system. Whenever funds were needed, the government began issuing Treasury bonds, and the Federal Reserve printed money to purchase these bonds, leading to a rapid increase in the money supply. Government debt soared from $391 billion (34% of GDP) in 1971 to $34 trillion (120% of GDP) by the end of 2023. During the financial crises of 2008 and 2020, the government amassed significant debt through this mechanism, resulting in the continuous depreciation of the dollar's value.

How long can such massive government debt be sustained? This question opens the door to various scenarios. One possibility is the emergence of another inflation fighter like Paul Volcker, who might take drastic measures to reduce debt, even at the cost of severe economic recession. Alternatively, disruptive innovations like artificial intelligence could boost supply and production, exerting continuous deflationary pressure on the economy and thereby extending the lifespan of the dollar.

However, as previously mentioned, currency is a social contract. Thus, the decline of the dollar will commence when the international community begins to lose faith in the United States and its currency. The inevitable inflation associated with being a reserve currency can exacerbate social issues such as income inequality and political polarization, both domestically and internationally, further eroding trust in the dollar. Although there are no definitive signs of the dollar's demise yet, accumulating issues suggest that such a scenario is increasingly plausible.

Geopolitical issues, not just inflation, can also undermine the dollar's status. In response to Russia's invasion of Ukraine, Western nations excluded Russia from the SWIFT banking system, preventing it from settling trade in euros or dollars. They also froze half of Russia's foreign exchange reserves held in dollars. Such actions can diminish other nations' trust in the dollar. For instance, China has been steadily selling off U.S. Treasury bonds and accumulating gold since the onset of the Russia-Ukraine conflict, thereby reducing its dependence on the U.S.

History proves that the dynamics of power surrounding currency remain constant. Unless an unprecedented perfect monetary policy emerges, any reserve currency will eventually lose its status. Although no one can predict the exact timing, the dollar will someday face its end. I can only hope that this moment comes as late and as smoothly as possible.

1.2 Bitcoin as a Hard Currency

As the dollar gradually loses its credibility, naturally, assets like gold, will garner attention. Gold has been valued from ancient times to the modern era due to its scarcity and immutable physical properties. During major conflicts, gold has been the ultimate asset recognized for its value internationally. Consequently, central banks around the world always maintain a certain reserve of gold.

Today, individuals can invest in gold through various means such as mining company stocks, gold futures, and gold ETFs. These investment methods are generally effective in developed countries with accessible financial markets. However, if you reside in a nation with less developed financial markets or one directly involved in war or revolution, investing in gold can be highly restrictive. These investment avenues do not involve direct ownership of gold, introducing counterparty risk during international turmoil. Additionally, purchasing and storing physical gold is not an easy task.

In such scenarios, Bitcoin can serve as an excellent hard asset similar to gold. Its supply is limited, it is not controlled by any single entity, and it is exceptionally easy to store and transfer, even in dire situations like wartime. For example, during Russia's invasion of Ukraine on February 24, 2022, the trading volume and price of BTC/UAH surged, trading at a 6% premium over the international rate. Even in less extreme cases, demand for Bitcoin is high in countries with unstable national currencies. In Turkey, where the annual inflation rate is around 70%, Bitcoin trades at a premium similar to gold. These examples demonstrate that Bitcoin can indeed fulfill the role of a hard asset.

Given the above examples, it is evident that Bitcoin holds significant potential to serve as a hard currency in the future. But does this mean that citizens of developed countries, currently protected by stable monetary systems, have no need to include Bitcoin in their portfolios? Even outside of crisis situations, allocating a portion of one's portfolio to Bitcoin can offer substantial benefits in terms of diversification. As illustrated in the graph, although Bitcoin's correlation with other assets like gold, stocks, and the dollar can be volatile over time, it generally exhibits distinct price movements. This unique characteristic alone makes it advantageous to hold a portion of assets in cryptocurrencies like Bitcoin.

Indeed, many financial institutions in the United States have recently added BTC ETFs to their portfolios. According to K33 Research, in the first quarter of 2024, 937 institutions reported holding Bitcoin ETFs in their 13F filings. Among them were notable names like JP Morgan, UBS, and Wells Fargo, as well as the Wisconsin Investment Board, which acquired BTC ETFs worth approximately $160 million. This trend indicates that Bitcoin is increasingly being recognized as a store of value.

Even before the inflationary effects of the COVID-19 era's quantitative easing have fully dissipated, the United States is increasing liquidity again in anticipation of the upcoming presidential election. The Treasury Department is expanding fiscal spending, and starting May 29, plans to conduct bond buybacks for the first time in over twenty years. Simultaneously, the Federal Reserve is slowing the pace of quantitative tightening.

Consequently, the dollar will continue to face inflationary pressures and will be issued in large quantities during major economic downturns. Unless the United States maintains its leadership through continual innovation in military, scientific, and industrial fields, the value of the dollar is bound to decline over time. Conversely, this will naturally increase the attention and value of Bitcoin.

However, to achieve the same status as gold as a hard asset, Bitcoin faces a critical challenge: the security scale and profitability of its network. The essential element for maintaining Bitcoin's value is the security level of its network. The more miners there are to mine Bitcoin, the more secure the network becomes, thereby solidifying Bitcoin's value.

Bitcoin miners earn revenue in two primary ways: block rewards and transaction fees. Block rewards are the Bitcoins awarded for successfully mining a block, with the amount being fixed and halved every four years. Transaction fees, on the other hand, are the fees paid by users for conducting transactions on the Bitcoin network, separate from block rewards.

For miners to continue participating in the Bitcoin network, their mining revenue must exceed their costs. Due to the halving that occurs every four years, block rewards diminish over time, necessitating an increase in transaction fee revenue to make up the difference. However, unlike networks such as Ethereum and Solana, the Bitcoin network has limited applications and low scalability, leading to fewer transactions and consequently lower transaction fee revenue. Recently, new token standards like Ordinals and Runes have momentarily increased activity on the Bitcoin network, but there is no long-term guarantee that these will significantly contribute to transaction fee revenue.

Up to now, mining revenue has generally surpassed mining costs. However, as block rewards continue to decrease due to future halvings, unless 1) Bitcoin's price rises substantially or 2) network activity increases to boost transaction fee revenue, there is a risk that miners will exit the network. This would lower the security level of the Bitcoin network, diminishing its intrinsic value and potentially leading to a vicious cycle of further miner departures and decreased security.

This highlights the primary difference between gold and Bitcoin. Gold's intrinsic value is not tied to profitability, whereas Bitcoin's intrinsic value is directly linked to it. Therefore, ensuring profitability is a long-term challenge that the Bitcoin network must address. While there is currently no definitive solution within the Bitcoin community, the emergence of applications such as Ordinals, Runes, and innovations like OP_CAT suggest a potential increase in transaction fee revenue in the long term.

2. Unlike Ever Before: AI

2.1 Impacts of AGI on Humanity

Historically, unlike currency, innovative technologies such as AI have always brought significant changes to society. The steam engine, electricity, and the internet revolution transformed the global industrial landscape, profoundly impacting human jobs and lifestyles. While these technological revolutions brought about various social issues during their transitional periods, they ultimately provided humans with much more prosperous lives. Steam engines and electricity liberated humans from most physical labor, while digital and internet technologies freed them from simple forms of mental labor.

AI technology has been studied since the 1900s, but meaningful results were slow to emerge. However, the pace of AI development accelerated dramatically after the publication of the Attention Is All You Need paper in 2017, which introduced the transformer theory. This breakthrough made it easier to develop large language models (LLMs), bringing humanity a step closer to artificial general intelligence (AGI). Like previous industrial revolutions, the development of AGI is expected to lead to a significant increase in productivity and have a substantial societal impact. However, I believe the implications will differ significantly for several reasons.

First, AGI will free humans from almost all forms of labor. Previous industrial revolutions liberated humans from physical and simple mental labor, leading to a higher proportion of the population engaged in more sophisticated tasks. However, AGI can handle advanced mental labor, including artistic endeavors such as art and music. Coupled with advanced robotics, this means the areas in which humans can contribute to productivity will diminish significantly.

(The modern day Luddite movement?)

Of course, this does not mean all jobs will disappear. Even in the 21st century, a portion of the population is engaged in agriculture and fisheries, although the proportion is much lower than in the past. While most job types will remain with the advent of AGI, the number of people needed to perform them will drastically decrease. For instance, tasks that ten people currently handle could be managed by one person in the future, leading to a significant increase in the population unable to find employment. Notably, leading figures in AI, such as Elon Musk and Sam Altman, have argued that AI and robots will handle global productivity, resulting in widespread job loss for humans.

Some argue that efficiency could be maximized while maintaining the current employment levels, but this is a misconception. For this to happen, demand would need to increase proportionally with the significant boost in supply (productivity) provided by AGI. However, in most fields, this is not feasible. Job creation would have to occur in new areas beyond AGI's reach, but as mentioned earlier, AGI's capabilities extend beyond physical to mental tasks, making this unlikely.

Secondly, AI is inherently a highly centralizable technology. Even before achieving AGI, the AI industry has already become heavily centralized around big tech companies. This is due to the rapid advancement of AI technology. Since the introduction of the transformer theory, the size of language models has increased by a factor of 10^4 between 2018 and 2022. Consequently, there are significant technological disparities in the essential industries that constitute AI technology.

• Semiconductor Design: Unlike the balanced consumer GPU market, NVIDIA almost monopolizes the data center GPU market used for AI model training and inference. This dominance is partly due to NVIDIA's CUDA toolkit, which is widely used by AI developers. The demand for NVIDIA's H100 GPU has surged, leading to longer delivery cycles. Thanks to this position, NVIDIA enjoys an impressive 78% operating margin, and the upcoming release of the Blackwell GPU in late 2024 is expected to further solidify NVIDIA's dominance. Although companies like AMD Xilinx and Intel Altera are expanding their FPGA businesses and big tech firms like Microsoft, Google, and Meta are developing their own AI semiconductors (ASICs), these solutions are still immature compared to GPUs in terms of market readiness and sophistication.

• Semiconductor Manufacturing: The foundry industry, responsible for manufacturing designed semiconductors, also shows a significant imbalance. NVIDIA's A100 production requires a 7nm process, and the H100 requires a 4nm process. These sub-10nm processes are practically monopolized by TSMC, Samsung, and Intel, with the A100 and H100 primarily produced by TSMC. TSMC is committed to producing NVIDIA's H100 for at least the next three years, and due to various factors, the gap between the first and others ranks in the foundry industry is expected to remain wide.

• Computing Power: AI companies require vast amounts of computing power for training and inference processes. This necessitates numerous AI semiconductors like the H100, large data centers, and substantial electrical power. According to Huawei, AI data centers are expected to account for 13% of global power consumption and 6% of the carbon footprint by 2030. The costs are also considerable; as Jensen Huang noted in his keynote at NVIDIA GTC 2024, training the GPT-MoE-1.8T (GPT-4) model required 8,000 H100 GPUs and 90 days. Thus, due to the necessity of securing AI semiconductors and bearing significant power costs, centralization in this industry is inevitable. Cloud services such as AWS and Azure, which provide computing power based on H100, are also unavoidably centralized.

• AI Models: While some AI models, like Meta's Llama and Google's BERT, are open-source, many others are closed-source. Closed-source models like OpenAI's GPT and Anthropic's Claude generally offer systematic development and better customer support compared to open-source models, but their centralization brings disadvantages in terms of cost and transparency.

• Data: Training AI models like LLMs requires enormous datasets. Legal arrangements, such as Google's $60M annual contract to use Reddit's data, are in place, but there are also numerous lawsuits regarding unauthorized data usage for AI model training. This has heightened interest in data sovereignty.

In summary, centralization is inevitable in the AI industry, where achieving economies of scale is essential. As the AI industry becomes more centralized, several micro-level issues can arise, such as excessive corporate profit-seeking, unethical data use, single points of failure like server downtimes, and the opacity of AI models. On a macro level, we may face societal chaos as the line between humans and AI blurs, and many people lose their jobs. I believe that blockchain technology, which inherently pursues decentralization, can serve as an antithesis to AI, addressing the challenges associated with AI centralization. Let's explore how blockchain can be applied to the AI industry.

2.2 Blockchain Can Fix AI

Just as Satoshi Nakamoto introduced Bitcoin in 2008, advocating decentralization in response to the unchecked issuance of currency by central banks, blockchain technology can be utilized in various ways in the AI industry, where centralization trends are driven by economies of scale.

Among the five highly centralized elements mentioned earlier, semiconductor design and production require concentrated expertise and substantial manufacturing facilities, leaving little room for blockchain solutions. However, blockchain can be effectively applied in the fields of 'computing power,' 'AI models,' and 'data.' Additionally, it can address issues such as the proliferation of fake information, including deepfakes, and support basic income policies for a populace facing mass unemployment. Let's explore the potential applications of blockchain technology within the AI pipeline.

Decentralized Computing

Training and inferring AI models require immense computing power and hardware. Big tech companies continuously purchase GPUs like NVIDIA's H100 for their model training, exacerbating the global hardware supply shortage. While services like AWS and Azure provide data centers for cloud-based AI model training and inference, they operate as oligopolies, imposing high margins on users. In response to these challenges, new services leveraging blockchain technology to offer decentralized computing power have emerged.

Examples include Akash and io.net, where users can contribute their hardware's computing power to the platform in exchange for incentives. There are also protocols specialized in niche services. For instance, Gensyn is optimized for training AI models. General decentralized computing services can reduce costs by utilizing idle hardware, but it is challenging to perform state-dependent computations, such as AI model training, in a decentralized manner. Gensyn addresses this with concepts like probabilistic proof-of-learning and graph-based pinpoint protocol. While Gensyn is specialized in training AI models, Bittensor focuses on AI model inference. Users can submit tasks, and Bittensor's decentralized nodes compete to provide the optimal results.

zkML

zkML, a fusion of zero-knowledge (zk) cryptography and machine learning (ML), promises to enhance the privacy and transparency of AI models. Many AI models currently operate as closed-source, leaving users uncertain whether these models are using the correct weights and performing inference honestly. By applying cryptographic techniques like ZK-SNARKs (Zero-Knowledge Succinct Non-Interactive Argument of Knowledge) to ML models, it becomes possible to prove that an AI model has executed its inference process correctly without revealing its weights, thus achieving both privacy and computational integrity.

ZK-SNARKs are a powerful cryptographic technology that allows the validity of arbitrary computations to be proven without revealing the input data. To illustrate this, consider a real-world example: proving one’s age online. Typically, this requires complex KYC verification, involving the disclosure of personal information such as name and ID. With ZK technology, this process can be simplified and made more private. Once a user has verified their age with an official entity, they can generate and submit a ZK proof whenever they need to prove they are over 18. This proof contains no personal information but still assures the verifier of the user's age, making the identity verification process safer and simpler.

Applying the same concept to ML models, a consumer using a closed-source ML model cannot be sure whether the model performed the computation honestly on the given input. By incorporating ZK-SNARKs, an ML provider can assure the consumer that the computation was carried out correctly without revealing the input or weights. A ZKP (Zero-Knowledge Proof) of the ML inference process can be generated and verified by a smart contract on a neutral blockchain protocol, ensuring that anyone can trust the results.

While the concept of zkML is highly attractive, significant challenges remain. Verifying ZKPs for specific computations is straightforward, but generating these proofs requires more computational power than performing the actual computation. According to Modulus Labs, generating a Plonky2-based ZKP for an ML model with 18 million parameters takes about one minute. Given that GPT-3 has 175 billion parameters and GPT-4 has 1.76 trillion parameters, substantial advancements are needed before zkML can be adopted meaningfully.

Data Sovereignty

As the AI industry continues to evolve, the significance of data grows exponentially. However, this surge has led to increasing instances of data sovereignty infringements. By leveraging blockchain technology, individuals can manage their identity-related information through self-custody, providing data only when necessary via digital signatures. Moreover, blockchain enables transparent data provision or sale through incentive systems or marketplaces accessible to all. Perhaps the most blockchain-like approach to data sovereignty has been exhibited by Reddit, which offered long-time users the chance to participate in its IPO, while contracting to provide data to Google. This move exemplifies a novel path in data sovereignty.

While slightly tangential to data sovereignty, blockchain also holds the potential to address issues in the data labeling industry. Data labeling is essential for enhancing the accuracy and ethics of AI models. Currently, this task often falls to low-wage workers, emerging as a new social issue. For instance, China's AI industry exploits vocational school students, and OpenAI has outsourced this work to low-wage workers in Kenya. Integrating blockchain into data labeling could democratize participation and ensure fair compensation.

Proof of Personhood

Decentralized computing, zkML, and data sovereignty may solve some AI industry challenges. Yet, proof of personhood and universal basic income could safeguard human sovereignty in a society drastically altered by AGI. Let us explore how blockchain might support human sovereignty amidst such profound social transformation.

As AI models advance, the production of various content forms—text, images, videos—by AI becomes increasingly prevalent. Distinguishing whether these outputs are human-made is becoming more challenging. The acceleration of digitalization is inevitable, and as AI-generated content proliferates, the associated social problems will undoubtedly surge.

These issues are not merely speculative; they are already occurring. Fraud through deepfakes, which mimic the faces and voices of individuals, has become alarmingly frequent, resulting in substantial financial losses. The authenticity of videos is now often hotly debated online due to the existence of deepfakes.

A recent incident involving Caitlyn Jenner illustrates this point vividly. She announced the launch of a meme coin on the Solana network via the platform X. Given the unusual nature of the announcement, many suspected her account had been hacked. Despite Caitlyn posting a video herself, there was significant controversy over whether it was a deepfake. This debate persisted until Caitlyn's manager also released a video, helping to somewhat settle the matter.

As we advance into the AI age, one of the most critical challenges will be proving one's humanity in the digital realm. This concept, known as "proof of personhood," aims to prevent sybil attacks and disinformation in the digital world. Currently, most applications rely on government-issued identity systems like passports or credit cards to verify personhood. However, these methods pose privacy risks and the potential for single points of failure. Thus, a truly digital identity system is essential. Blockchain technology offers a solution, allowing individuals to prove their humanity and the authenticity of their created content, potentially mitigating issues like deepfakes.

The most commonly used method for digital identity verification is biometric systems, which authenticate specific body parts. OpenAI's CEO Sam Altman is pioneering a project called Worldcoin, combining blockchain technology with iris scanning. Users install an application on their mobile devices, receiving a private key (account) on the blockchain. By using an iris scanning device called the Orb, users can authenticate their humanity in the digital world. The Orb ensures that the user is indeed a person and that the iris has not been previously registered, securely granting digital identity.

The Orb transmits only the hash value of the iris data to the server, destroying the actual iris data afterward. Users can later prove their personhood without revealing their account address, thanks to ZK-SNARKs, addressing privacy concerns. However, potential issues like hardware backdoors still need to be resolved. The importance of proof of personhood extends beyond content authenticity. It plays a crucial role in the concept of universal basic income, which we will explore next.

Universal Basic Income

As previously mentioned, the advent of AGI is poised to bring about an unprecedented leap in productivity in human history. However, this revolutionary progress will inevitably result in significant job displacement. To sustain societal stability, the concept and necessity of Universal Basic Income (UBI) are gaining increasing attention. The idea of UBI predates AGI, tracing its origins back to Thomas More's "Utopia" in the 16th century. UBI entails providing regular, unconditional financial support to all members of society. An existing example of UBI can be found in Alaska, where the Alaska Permanent Fund Dividend offers a form of UBI, demonstrating positive outcomes across various dimensions such as poverty, employment, and health.

The focus here, however, is not on a UBI that merely enhances quality of life, but on a UBI substantial enough to support individuals who lose their jobs due to AGI, ensuring they can live adequately without employment. Elon Musk refers to this as "universal high income." Similarly, Sam Altman has shown considerable interest in UBI, conducting research through OpenResearch. He has proposed innovative ideas such as providing UBI in the form of assets and means of production like equity or computing power, rather than just cash.

Sam Altman's Worldcoin, discussed in the "Proof of Personhood" section, is also closely linked to UBI. A critical aspect of UBI distribution is ensuring that only genuine individuals receive it and preventing multiple claims by the same person. Thus, preventing Sybil attacks is crucial for implementing UBI. Worldcoin aims to achieve this through iris recognition for proof of personhood. Currently, users verified via iris recognition on the Worldcoin app receive WLD tokens periodically, a form of UBI. Although I resonate with Worldcoin's vision, I harbor some reservations about the distribution of WLD tokens.

Even beyond Sam Altman's Worldcoin, blockchain technology will be indispensable for establishing a complete UBI system. Blockchain can enhance transparency and efficiency not only in selecting recipients through proof of personhood but also in the distribution process, ensuring a more effective and transparent UBI delivery.

3. Anyway, Humanity Will Need Blockchain

Despite the unprecedented crises marked by the collapses of Terra and FTX, the blockchain market has swiftly regained its scale. However, reflecting on both the previous and current market booms, a distinct shift in the industry's vision is evident. In 2021, numerous protocols were driven by the grand vision of decentralization, capturing the imagination and excitement of many. Now, despite the market's similar scale, there seems to be a pervasive uncertainty within the industry and community about the direction blockchain should take. This is not due to any failure on our part or a deficiency in blockchain technology itself; rather, it is simply that the current era has not yet created a pressing need for blockchain technology.

While it is intriguing to observe blockchain's application in niche markets, the industry must set its sights higher. As the long history of humanity has shown, we will continue to experience cyclical monetary systems and revolutionary technological innovations. Within these vast movements, blockchain will stand as a crucial technology that will safeguard human sovereignty.