Facta non verba. https://paragraph.com/@danpark Wisdom for action. Sun, 15 Nov 2026 12:48:15 GMT https://validator.w3.org/feed/docs/rss2.html https://github.com/jpmonette/feed en Facta non verba. https://storage.googleapis.com/papyrus_images/0eb3eb05e13f6b90cf8d40bad561dd6f https://paragraph.com/@danpark All rights reserved <![CDATA[Facta non verba. Edition 1.]]> https://paragraph.com/@danpark/facta-non-verba-edition-1 NGiAGPEpMYtzStKKp0Ss Sun, 09 Apr 2023 13:52:04 GMT Hey folks! Welcome to the Facta non verba. newsletter. I'm thrilled to announce the first edition of my newsletter today! While I can't promise a set schedule, my aim is to deliver at least two editions per month, keeping you updated on the market and sharing my personal insights. Additionally, I plan to share some original writing that doesn't quite fit the newsletter mold.

It took me a while to find the perfect platform for sharing my thoughts, especially since I plan to do this for as long as possible (fingers crossed my schedule permits). After weighing the pros and cons of traditional services like Substack and up-and-coming platforms like Beehiiv and Ghost, I finally settled on Paragraph.xyz. Here's why I believe Paragraph.xyz is the right choice for me, broken down into three key factors:

  1. Web3 features like wallet-based subscriptions and token-gated views are right up my alley, considering my involvement in the crypto & blockchain industry. This should help me connect with potential subscribers who share the same interests right from the get-go.

  2. The Paragraph.xyz team strikes a great balance between a sleek Web2 user interface and the robust engagement potential offered by Web3. While there have been some notable failures in the "Web3 blogging platform" arena (RIP Steemit), I believe this time could be different.

  3. Choosing a smaller platform like Paragraph.xyz allows me to stand out in the saturated world of content creation (does that sound a tad hippie?). Sure, it might be a bit of a wild bet, but if my "Web3 blogging will succeed at last" thesis holds true, I stand to reap the rewards of my gamble.

In case you weren't aware, I also run a public Telegram channel named "Facta non verba." where I share my daily memos. This helps me keep track of my insights and connect the dots more effectively. If you haven't joined already, feel free to hop on board.

Alright, without further ado, let's dive in!

Intro

I go to school these days. As an architectural engineering major deeply into other stuffs (which some of you already know), I might not be the most active student in most of my classes, but there's one lecture I genuinely enjoy. It's the one that elaborates how the development of vertical software is interconnected.

During the first hour, we delved into the history of CAD. It's fascinating to learn that CAD was invented back in the 1960s, long before the age of the mouse, and it took quite some time for it to evolve into a practical product. ArchiCAD emerged from a Hungarian team, while Autodesk introduced its CAD in the late 80s and 90s alongside the commercialization of Mac and Windows. The turning point came with the advent of personal computers.

Surprisingly, the AEC industry has also adapted to the mobile era, which was dominated by the iPhone. Companies like Procore led the way with mobile apps for project scheduling and management. Admittedly, they were a bit clunky on mobile and better suited for tablets. (I remember using a Sketchup-like tool on my iPad during my freshman year, but the entire experience wasn't that great to be honest.)

The point is, even an industry as slow and conservative as AEC can't help but respond to major shifts. But does that always mean a change in leadership? Not necessarily. Despite asynchronous workflows and a quirky UI, I always thought we'd witness the rise of a Figma for AEC or a cloud-based BIM upstart. However, Autodesk remains king.

Anyway, what's convinced me that the AI wave is the real deal is the growing opportunity I've noticed even within the AEC industry (= which is v conservative).

Big Figures Speaking of AI

Sam Altman

Sam Altman: OpenAI CEO on GPT-4, ChatGPT, and the Future of AI | Lex Fridman Podcast #367

The hottest man on the planet speaking. Let's skip the comments that are easy to find elsewhere and dive right into the highlights that truly stood out (at least for me).

Sam's thoughts on the societal changes driven by AGI and the necessity of UBI (1:39:25)

Lex Fridman: You have been a proponent of UBI, Universal Basic Income. In the context of AI, can you describe your philosophy there of our human future with UBI? Why you like it? What are some limitations?

Sam Altman: I think it is a component of something we should pursue. It is not a full solution. I think people work for lots of reasons besides money. And I think we are gonna find incredible new jobs and society, as a whole, and people as individuals, are gonna get much, much richer. But, as a cushion through a dramatic transition, and as just like, you know, I think the world should eliminate poverty if able to do so. I think it's a great thing to do as a small part of the bucket of solutions. I helped a startup project called Worldcoin, which is a technological solution to this. We also have funded a large, I think maybe the largest and most comprehensive universal basic income study as part of sponsored by OpenAI. And I think it's, like, an area we should just be looking into.

And his holistic approach to addressing the myriad of events unfolding in the present day (2:20:34)

Lex: What do you think is the meaning of this whole thing? That's a question you could ask an AGI. What's the meaning of life? As far as you look at it? You're part of a small group of people that are creating something truly special. Something that feels like, almost feels like humanity was always moving towards.

Sam: Yeah, that's what I was gonna say is I don't think it's a small group of people. I think this is, like, the product of the culmination of whatever you want to call it, an amazing amount of human effort. And if you think about everything that had to come together for this to happen. When those people discovered the transistor in the 40's, like, is this what they were planning on? All of the work, the hundreds of thousands, millions of people, whatever it's been, that it took to go from that one first transistor to packing the numbers we do into a chip and figuring out how to wire them all up together and everything else that goes into this. You know, the energy required, the science, like, just every step. Like, this is the output of, like, all of us. And I think that's pretty cool.

The Contradictions of Sam Altman, AI Crusader

A good article on Sam Altman's role in transforming OpenAI from a nonprofit to a multibillion-dollar company. The article elaborates the controversies and criticism surrounding OpenAI's commercialization and partnership with Microsoft.

And one important point to think about: why did they brought ChatGPT now? Maybe,

1) they were determined that they were leading the game,

2) they predicted that the GenAI-based product will spread out very fast, and

3) owning that workflow by spreading out (unfinished) product would provide them a huge moat.

Now it is clear that they're leading the game, and the AI safety is largely up to them. Sam probably wanted to create a circumstance like this.

In its founding charter, OpenAI pledged to abandon its research efforts if another project came close to building AGI before it did. The goal, the company said, was to avoid a race toward building dangerous AI systems fueled by competition and instead prioritize the safety of humanity.

And he also thought of launching a cryptocurrency for ChatGPT:

That year, Mr. Altman said he looked into options to raise more money for OpenAI, such as securing federal funding and launching a new cryptocurrency. “No one wanted to fund this in any way,” he said. “It was a really hard time.”

Sam Altman - How to Succeed with a Startup

2:00~3:30 is my favorite part that I revisit at least twice a year. Here's the entire transcript:

A really important thing to figure this out is learning how to differentiate between real trends and fake trends. A real trend is something that's actually going to happen in the fake trend is not or at least not yet. And before you make a big bet on a new platform, you want to make sure it's real. Now there's an easy trick for this which I'll share now.

Real trends are ones where a new technology platform comes along and the early adopters use it obsessively and tell the friends how much they love it. A fake trend is one where people may buy the product but don't use it or at least not enough.

So an example of a real trend (I already mentioned the iPhone I'll mention that again) when the iPhone first came out many people were dismissive because they only sold a million or two million that year, and they said well this just doesn't matter. But for the people that had an iPhone they used it for hours every day, it became central to their lives, they loved it they told their friends you've got to get one. And I think it was obvious then do people pay an attention that something had fundamentally shifted and we had a new computing platform that was gonna spawn huge businesses, and it was a good time to bet on mobile apps.

A fake trend or at least a fake trend as of August 2018, I would say as VR. I do believe VR will be big someday, but today most people that I know that own a VR headset use it never or very rarely. And so although a lot of people talk about it and maybe even a lot of people buy them there's not the intense usage per user among the early adopters that I think you want to see, before you make a big bet.

If I apply ChatGPT as an alternative for iPhone... it looks quite same. He's out there building the exact same thing he has mentioned.

Bill Gates

Read his book <The Road Ahead> for the last weekend. A bold declaration towards the 'information superhighway,' and the relentless efforts he had shown to seize the once-in-a-lifetime opportunity, were really thought-provoking.

The Age of AI has begun

And then I re-read his recent article on AI. For this time, I went for the entire article without ChatGPT support. Think we should listen carefully to rare figures like him. Here's some of the lines I've highlited:

I knew I had just seen the most important advance in technology since the graphical user interface.

This inspired me to think about all the things that AI can achieve in the next five to 10 years.

The development of AI is as fundamental as the creation of the microprocessor, the personal computer, the Internet, and the mobile phone. It will change the way people work, learn, travel, get health care, and communicate with each other. Entire industries will reorient around it. Businesses will distinguish themselves by how well they use it.

I knew I had just seen the most important advance in technology since the graphical user interface.

The biggest evolution throughout software history has always been better technology improving user experience. Nobody hardly wants to enter hard and complicated interfaces.

Elad Gil and Reid Hoffman

And if you've got a minute, I'd suggest taking a look at this short video as well.

Recent Papers, Concerns, and the Pros and Cons of Rapid AI Development

GPT-related papers

Sparks of Artificial General Intelligence: Early experiments with GPT-4

Wow. Microsoft Research mentioned this: "We believe that it could reasonably be viewed as an early (yet still incomplete) version of an artificial general intelligence (AGI) system."

  • GPT-4 part of new cohort of LLMs with more general intelligence; excels in various domains without special prompting.

  • GPT-4 could be seen as early, incomplete version of AGI.

  • Possible need for new paradigm beyond next-word prediction.

GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models

  • Investigating implications of GPT models on the U.S. labor market; new rubric to assess occupations based on GPT capabilities.

  • 80% of workforce may have at least 10% of tasks affected by GPTs and 19% of workers may see at least 50% of tasks impacted.

  • Influence spans all wage levels, higher-income jobs may face greater exposure; impact not limited to industries with higher recent productivity growth.

For those who don't have much time to read — recommend this thread by Jim Fan:

BloombergGPT: A Large Language Model for Finance

BloombergGPT is a 50B-parameter model specialized in the financial domain. It's trained on a vast dataset of 363B tokens from Bloomberg and 345B tokens from general-purpose datasets.

This model outperforms existing models on financial tasks without compromising general LLM benchmark performance. Future plans include releasing detailed training logs.

2023 AI Index Report by Stanford University

Didn't have much time to go over either, so I'd recommend the thread below for better understanding:

And some of the graphics that caught my eye:

You should be in the US if you're starting an AI company.
... and people in China believe in AI's future way more than people in the US.

AI and Copyrights

One thing I notice these days: people tend to rely on comfort until things become really serious. Even I don't care what & how OpenAI specify and track my data and the possibility of potential use as for-profit ways. — ChatGPT is a data privacy nightmare. If you’ve ever posted online, you ought to be concerned

Adobe made an AI image generator — and says it didn’t steal artists’ work to do it

Indeed, finding the proper way to compensate the people who helped training the specific data model will be a v important problem for upcoming decades. Adobe clearly shows how each domain will serve the data providers.

Adobe says their AI models are trained on content that is either out of copyright, licensed, or from the Adobe Stock library and intends to pay artists who contribute training data, with a compensation strategy currently under development.

To make its AI tools more user-friendly, Adobe includes built-in options for art styles, lighting, and aspect ratio. A future Photoshop feature will allow artists to train the system on their own work to generate content in their personal style, though potential abuse of this feature is a concern.

Copy-protection tech would grow way much bigger along with the compensation strategy. Adobe is working on a "Do Not Train" system that enables artists to block AI from training on their work and is in conversation with other model creators to collaborate on this standard.

Pause Giant AI Experiments: An Open Letter — and the Oppositions

We all know about what's going on right now; and the letter below addresses a much larger issue than just copyright:

And I could see some quite differentiated opinion compared to it. Yann LeCun says:

There are a lot of issues with making AI systems controllable, and making them factual, if they’re supposed to provide information, etc., and making them non-toxic. There is a bit of a lack of imagination in the sense of, it’s not like future AI systems will be designed on the same blueprint as current auto-regressive LLMs like ChatGPT and GPT-4 or other systems before them like Galactica or Bard or whatever. I think there’s going to be new ideas that are gonna make those systems much more controllable. — from Titans of AI industry Andrew Ng and Yann LeCun oppose call for pause on powerful AI systems

... and Albert Wenger is also against it:

"We should still think about other regulation to address structural risks but much of what has been proposed here doesn’t make a ton of sense. For example, publishing an algorithm isn’t that helpful if you don’t also publish all the data running through it. In the case of a neural network alternatively you could require publishing the network structure and weights but that would be tantamount to open sourcing the entire model as now anyone could replicate it. So for now I believe the focus of regulation should be avoiding a situation where there are just a few huge models that have a ton of market power." — from Thinking About AI: Part 2 - Structural Risks

"Bottom line then for me is that “loss of reality” is an existential threat, but one that we have already been facing and where further AI advancement will both help and hurt. So I am not losing any sleep over it. There is, however, an overlap with a second type of existential risk, which is a super intelligence simply wiping out humanity. The overlap is that the AI could be using the loss of reality to accomplish its goals." — from Thinking About AI: Part 3 - Existential Risk (Loss of Reality)

Recent Developments

ChatGPT Plugins

This was to be honest, mind-blowing. Packy well elaborates what this 'Plug-in play' for means for OpenAI in his recent article "Attention is All You Need."

Strong API-first businesses sit in this sweet spot: they provide mission critical but non-core functionality to their customers, like accepting payments, providing cloud security, or sending communications to customers.

OpenAI has even expanded its sales operations, hiring experienced sales leaders and providing technical support to customers. And the important thing is that the customers (=Plug-in partners), are willing to pay because the product deserves it. The rapid and robust construction of their ecosystems (with steep revenue growth — that is even fully under OpenAI's control) is truly remarkable.

If you wonder what it means for Microsoft, I recommend you reading "The Accidental Consumer Tech Company; ChatGPT, Meta, and Product-Market Fit; Aggregation and APIs" by Ben Thompson. The deal may not have been as favorable to Microsoft as you think.

On the flip side, I think that ChatGPT + plugins means that Bing might not make the ultimate market share progress that Microsoft hopes: while OpenAI withdrew the web plugin at some point over the weekend, I’m sure it’s coming back, and the reality is that in the consumer space ChatGPT already has the mindshare and, in my estimation, the better user experience.

Of course we have seen clearly compatible companies fight on each other’s turf to their collective detriment (e.g. Apple and Google), but the fact that Microsoft and OpenAI are joined at the hip will perhaps make their coopetition tilt more towards the “cooperate” part than the “competition” part. That’s for them to decide, though; it’s customers who are the big winners, which is always the origin story of a new Aggregator.

Segment Anything

Meta plans to commercialize its generative AI by December, targeting improved advertising effectiveness and integration into its products and services, including the metaverse. More can be found on: "Meta to debut ad-creating generative AI this year, CTO says."

And FAIR recently announced Segment Anything. Yann LeCun had a reason for being so confident these days.

Introducing Segment Anything: Working toward the first foundation model for image segmentation

  • The Segment Anything introduces SAM, a flexible image segmentation model that can adapt to different tasks without extra training. It simplifies the segmentation process and reduces the need for specialized expertise.

  • The project also presents the largest-ever segmentation dataset, SA-1B, which can help improve future models. SAM has potential applications in AI research, AR/VR, content creation, and scientific study, making it a significant advancement in computer vision.

And Jim Fan's takeaway: "Reading Meta's Segment-Anything, and I believe today is one of the "GPT-3 moments" in computer vision."

Runway

And TTV(Text-to-Video) is coming to life. VTV(Video-to-Video) already live at runwayml.com.

Runway Pricing Plans

But the pricing plan's not so generous for (almost) all of us. More to go on the video generation side.

Runway AI Film Festival

For a deeper understanding of how things truly function in real, recommend viewing some of the winning films.

Building AI Apps, in a B2B Scale

To be honest, I don't think AI is causing any major shifts at the moment. Sure, consumer-facing businesses have changed a lot since the introduction of GUI, as Bill Gates pointed out. But it's the behind-the-scenes workflows that have seen the most significant impact. Right now, we're just seeing cool new interfaces and some of end-user experiences. The real game-changer for the entire workplace is still on the horizon.

For B2B Generative AI Apps, Is Less More?

Similar approach from a16z's recent article. Wave 2 of GenAI (= SynthAI) focuses on synthesizing and converging information to improve decision-making in B2B apps. It aims to enhance the quality and speed of decision-making by summarizing vast amounts of information that humans cannot process themselves.

SynthAI's real value lies in helping humans make better decisions faster and rethinking the UX to convey information efficiently. Examples of SynthAI applications include AI-powered knowledge bases and identifying target accounts at their highest intent level.

To ensure high-quality synthesis, there will be a shift from large-scale, generic models to architectures that leverage multiple models, including fine-tuned models for domain-specific datasets. This enables the creation of moats in the delivery of speed and quality.

Wave 2 will benefit most from SynthAI when there is a high volume of information and a high signal-to-noise ratio.

The battle between existing systems and new AI-native solutions is centered on owning the workflow. Existing solutions aim to improve their workflows with AI, while challengers use best-in-class AI implementations as a wedge to redefine the workflow.

“In the future, we think there’s an opportunity to allow the user to ask follow-up questions if they want to dig further into a theme. At the end of the day, it’s about delivering the end-to-end workflow — from gathering data quickly to understanding it quickly — to help make decisions in real time.” — by Ryan Glasgow, Sprig Founder & CEO

And might be kinda off topic but have a gut feeling that there will be a lot of uncertainty surrounding the usage of LLM-based services especially in large corps. Recently met some friends working in conglomerates said that ChatGPT and similar tools are blocked by their internal firewalls. Think LLMs will become more verticalized, with a much higher security bar set for B2B sales (and well every big company are likely to have their own models).

Crypto Should Learn Something from AI

Crypto vs. AI?

I love John Luttig's writings. His latest writing put AI and crypto to the table with 4 different aspects. And even I'm more into crypto, I could hardly disagree with his viewpoints. Think his words hint the pitfalls of crypto and how the industry should evolve in a longer-perspective.

Is AI the new crypto?

1. Capital: Crypto faced capital challenges due to a mismatch between capital inflows and progress, as well as VC short-termism, leading to inflated valuations and misaligned incentives. In contrast, AI is less likely to experience these challenges, as generalist VC firms will maintain a balanced approach and focus on specific applications.

2. Mission: Crypto's founding mission was anarcho-capitalist, aiming to disrupt financial institutions, but it became less clear as centralization increased. AI, however, lacks a clear founding mission and should aim for neutrality at the infrastructure level, with applications defining their specific missions.

3. People: Crypto initially attracted libertarians and anarchists but later drew in grifters, leading to a hype-driven culture and a market bubble. AI's talent evolution arc is purer, with a research-focused culture that prioritizes genuine technologists while navigating regulatory risks and political controversies.

4. Value creation: Crypto's value creation is unclear and still in the infrastructure phase, with stalled adoption in many commercial dimensions. AI demonstrates clear value creation through various applications and is expected to capture value through infrastructure oligopolies and widespread application proliferation.

Learning from AI

Indeed, people want their life to become easier, not harder. Simple and easy. Same for blockchain if we really believe in & want mass adoption. In my view, crypto is not as big because it is oriented to the tech itself, not the fundamental shift of user experience. Reminds me of SISP: Solution In Search of a Problem.

From Kevin Hale - How to Evaluate Startup Ideas

But remember, mobile apps fundamentally have changed nearly almost everything on the internet: from finance, payments, contents, social, commerce, productivity, and whatnots.

The important thing is, whether the orientation of crypto industry as a whole goes well with the NBT. As always, the biggest portion of the value accrue will come from the consumer-facing layer. I don't think crypto and AI is not just like oil and water. They will go well together. Crypto can be a layer sitting atop AI, working as the next layer of finance, payments, contents, and whatnots.

And my short thoughts on the 'investible' and 'x-to-earn' aspect of crypto; think it could and should be a part of it, but if it goes beyond the main agenda of the industry as a whole, something might be broken.

The mass always wants to PAY money for something that benefits them a lot, and only a small number of ppl actually want to EARN money. And this 'earning' part should accompany a set of skills or hardship that somehow benefits other participants, and incentives for those activities must be considered v carefully, because the incentive is always comparative.

Targeting Crypto-natives v. Normies — a Thread

I admit that the approach of targeting crypto users vs. internet users is valid because one-time crypto users are highly likely to participate more than once due to the high barrier and stickiness, which makes it an irreversible experience as internet showed. But I don't think this experience (of buying crypto and playing around with them) converts to a powerful daily engagement.

Only a small portion of crypto projects, which are the 'crypto-native projects', can achieve daily-level engagements with less than 1 million users. I am not trying to say that these projects are meaningless, but the order of magnitude is still small compared to what crypto can achieve.

Crypto Meets AI

Worldcoin

Recent article from Sam Altman's 'real' crypto project. Now 1.4M+ registrations made, and think Worldcoin can be one of the biggest cryptocurrency prj ever if successful.

Humanness in the Age of AI

In the age of advanced AI, proving humanness online has become increasingly difficult. Worldcoin's World ID protocol addresses this challenge through biometric-based proof of personhood (PoP), using a custom, open-source hardware device to validate liveness and uniqueness without storing image data. The project aims to create a secure and private foundation for digital identity, focusing on the first two layers of a three-layer system: PoP and digital authentication.

Proof of personhood is essential for mitigating AI-related challenges, such as sybil attacks and the spread of AI-generated content. A global PoP network can enable wealth redistribution through UBI, advanced spam filters, reputation systems, inclusive democratic governance models, strengthened authentication, and equitable distribution of scarce resources.

Key considerations for building PoP on a global scale include privacy, self-sovereignty, fraud resistance, inclusivity, scalability, decentralization, and continuity. Among potential PoP mechanisms, biometrics, specifically iris recognition, emerges as the most promising method. However, improvements in hardware and technology are required for secure and inclusive global implementation.

A good overview on how sufficient decentralization works for them — A Primer on Decentralization at Worldcoin

Planning for AGI and beyond

It is quite well aligned with mission statement of OpenAI. I feel OpenAI & Worldcoin could have a relationship somewhat similar to that of SpaceX & Tesla. Different entities under same umbrella.

OpenAI emphasizes the importance of gradually transitioning to a world with AGI, learning from deploying less powerful systems, and developing increasingly aligned and steerable models.

The short-term goals include 1) gaining experience in operating AI systems, 2) creating more aligned models, and 3) initiating a global conversation on governance, fair distribution of benefits, and fair access to AI systems (which would be possibly deeply connected with Worldcoin imo).

In the long term, the future of humanity should be determined by humanity itself, with public consultation for major decisions. The first AGI will be a point along the continuum of intelligence, and progress will likely continue, potentially leading to a vastly different world with extraordinary risks.

LLM for Blockchain?

Dune’s LLM Roadmap: Transforming the Way You Interact with Data

Eventually, it seemed like the data layer would be the first to adopt LLM, but Dune got there first. Recent Dune’s LLM Roadmap aims to revolutionize data interaction in the crypto data space by introducing a suite of AI-powered features. Their ultimate goal may finally be achievable by LLM. Rather than building a strong moat, it was more of a strategy to stay on top by moving one step ahead of everyone else. It's like being the first to launch an app in the mobile era.

The first feature, Query Explanations, demystifies complex SQL queries, making them more accessible. Upcoming features include: Query Translations to facilitate the transition to Dune SQL, NL Querying for seamless data analysis, Improved search functionality for finding relevant information, and Wizard Knowledge Base chatbot to help users navigate and contribute to existing resources.

These advancements underscore Dune's commitment to making crypto data more accessible and user-friendly for their community of Dune Wizards.

Swalo - The NL Chat-based Transaction Administrator

While Swalo.ai's competitive advantage may be limited, it's an excellent demonstration of agility. Ultimately, the struggle revolves around capturing share of user flow. Much like launching a mobile app in the mobile era wasn't a moat but a necessity for survival, one might speculate whether, in a few months, it will become essential for successful products to integrate features (like LLM for enhanced transaction simulators like security assurance and transaction comprehension).

The Give-to-Get Model

And David Sacks introduced a distinct perspective concerning domain-specific models, which could potentially forge a connection between the realms of crypto and AI.

The Give-to-Get Model for AI Startups

Give-to-get model, pioneered by Jigsaw, could be highly beneficial for AI startups that need rich proprietary datasets to train their models.

By incentivizing professionals in various industries to share their data, AI startups can rapidly train and improve their models to serve those professionals, leading to better performance, valuable insights, and a strong competitive advantage.

Give-to-get model for vertical industry has potential, and we hopefully may witness some decentralized compensation models arising. Adobe's recent compensation strategy (mentioned above) can be acknowledged as an early (but centralized) example for this.

Crypto Regulations

What's Happening in the US?

It is very clear that US lagging behind the others in regulatory perspective. It is now evident that France, UK, UAE, HK, SG, Australia, and Japan are rapidly recognizing the opportunities in crypto and creating a supportive environment whereas US is stuck with regulation by enforcement. — See "Europe is winning. Will the US catch up?" for details.

Exercise Caution with Crypto Asset Securities: Investor Alert

SEC recently posted an alert. So adversarial.

The SEC’s Office of Investor Education and Advocacy continues to urge investors to be cautious if considering an investment involving crypto asset securities. Investments in crypto asset securities can be exceptionally volatile and speculative, and the platforms where investors buy, sell, borrow, or lend these securities may lack important protections for investors.

The risk of loss for individual investors who participate in transactions involving crypto assets, including crypto asset securities, remains significant. The only money you should put at risk with any speculative investment is money you can afford to lose entirely.

To understand deeply on this agenda, highly recommend the Paradigm Policy's SEC's Path to Registration series. They have now released Part 2 out of the four-part series.

Part I. Due to SEC Inaction, Registration is Not a Viable Path for Crypto Projects

The SEC's inaction has made registration an unviable path for crypto projects. Gary Gensler has implied that registering tokens with the SEC is simple, but the reality is far more complex. The current registration forms are inadequate for addressing crypto's unique aspects, and the SEC has failed to provide a workable regulatory framework for security tokens.

e.g. Companies like Coinbase have struggled with the lack of clarity in the registration process. The SEC's "just come and register" stance makes no sense.

Part II. Lessons from Crypto Projects’ Failed Attempts to Register with the SEC

The SEC registration process for has proven to be ineffective due to various factors, leading to many projects failing or facing disadvantages in the market. The current regulatory framework lacks clear and workable rules for digital asset securities, resulting in limited infrastructure for trading registered tokens and high costs with compliance.

This is primarily due to the SEC's reluctance to provide a workable framework through rulemaking, exemptive relief, guidance, and industry engagement. Instead, the SEC has opted for a regulation by enforcement approach, which has created a challenging environment.

Yes, there was Reg A offerings, but companies that attempted Reg A token offerings, such as Hiro Systems PBC, YouNow, and Ceres Coin LLC, faced numerous challenges as well. These cases ultimately revealed the limitations of Reg A as a viable pathway for compliant token offerings.

* Regulation A (Reg A) offered a "slimmed-down" alternative to traditional IPOs for small companies. However, it did not turn out to be a viable option due to high costs, trading restrictions, and burdensome disclosure obligations.

White House blasts digital assets in new report, sees little value in crypto

In the meantime, White House released report criticizing digital assets and their purported benefits. The White House's stance on digital assets may be shifting from agnostic to openly adversarial as well.

TL;DR:

  • Report argues that crypto assets have not delivered on promises of better payment mechanisms, financial inclusion, or cutting out middlemen.

  • The report questions the role of regulation in protecting consumers and the financial system from crypto-related risks.

  • The report also doubts the potential of a CBDC but acknowledges possible benefits of DLT in the future. Document critiques stablecoins, calling them "too risky" for broad payments purposes.

  • The White House has a negative view of DLT and PoW.

  • Report highlights the risks associated with DeFi applications.

  • The administration's majority opinion on digital assets remains uncertain.

Centralized Exchange Tribulations

We asked the SEC for reasonable crypto rules for Americans. We got legal threats instead.

SEC issued a Wells notice to Coinbase (regarding listed digital assets, staking service, Coinbase Earn, Coinbase Prime, and Coinbase Wallet). It is not a formal charge or lawsuit but can lead to one; Coinbase products and services continue to operate as usual.

Coinbase provided multiple registration proposals to SEC, but received no response. They have met with SEC over 30 times in nine months, with no feedback on their proposals.

In the meantime, Coinbase is going more out of US with their 8-week international expansion plan. Another location to watch out for Coinbase's full-fledged entry. Coinbase integrated Pix for easy account top-up using Brazilian Reals, localized app with 24/7 email support, partnered with Ebanx, and streamlined the user onboarding process as a part of 8-week International sprint. — Introducing Exciting New Updates for our Brazilian Community

Binance argues that the civil complaint by CFTC contains incomplete facts and disagrees with many of the allegations. The company highlights its commitment to compliance & transparency and also stresses its extensive licensing, strict employee trading policies, and its intention to continue collaborating with regulators worldwide. — from CZ’s Response to the CFTC Complaint

7 details in the CFTC lawsuit against Binance you may have missed

  1. Tokens labeled as commodities: (The CFTC) labeled BTC, ETH, LTC, USDT, and BUSD as commodities, contrary to SEC's assertions

  2. CZ's phone accessed: gathered evidence from CZ's phone, collecting Signal text chains and group chats

  3. Terrorist activity accusations: accused Binance employees of knowing that its platform had facilitated potentially illegal activities

  4. One man at the top: alleged that CZ controlled dozens of entities operating the Binance platform as a common enterprise, with him personally approving minor expenses

  5. VIP program perks: scrutinized Binance's VIP program, alleging that VIP customers were given prompt notification of any law enforcement inquiry about their account

  6. Ignoring U.S. regulatory requirements: (Binance) was aware of US regulatory requirements but ignoring them and making deliberate decisions to evade federal law

  7. Fines and injunctions: (The CFTC) is seeking monetary penalties, disgorgement of profits, and a permanent injunction against further violations

One thing we always have to check - which comps & orgs will benefit if one competitor faces hardships? Ofc regulations by enforcement on CEXs are all difficult at the moment; Staking service for Kraken, Wells notice for Coinbase, and now Binance...

So I strongly agree that while the lawsuit and regulatory uncertainty in the U.S. present risks for companies operating in the country, jurisdictions like SG, UAE, and some European countries could emerge as the true winners. — from Binance's bad days could be good for Kraken

And for more statistical approach, recommend "CFTC v. Binance: The Data Behind the Suit"

Since the CFTC lawsuit, the exchange has seen $2.2B of crypto flow out, but also $1.3B sent in, resulting in a net outflow of $900M. Experts say there is no real sign of a bank run (as of Mar 29), as Binance holds $64B in on-chain assets, with 50% in stablecoins. — from Analysts say Binance’s $2.2 billion outflows don’t suggest a bank run yet

If you wanna track the live status — Binance Exchange's Health After the CFTC Lawsuit - Comparative analysis

Europe

MiCA's impact on the EU crypto industry and beyond

Europe is trying to solidify their lead with crypto regulation while US struggles.

Below is the TL;DR:

MiCA is a game-changer for the European crypto industry, providing a harmonized regulatory framework across all 27 countries. It will increase competitiveness and market share for regulated businesses, encouraging institutional adoption and potentially attracting global capital, talent, and companies.

However, MiCA's success depends on the practical implementation standards and enforcement practices to be developed by EU supervisory authorities over the next 12-18 months. If MiCA proves to be workable for the industry, consumers, and regulators, it will likely have a global impact similar to GDPR for privacy. This would make it the most comprehensive regulatory framework for crypto-assets seen on a global scale.

It is crucial for industry participants to engage closely with EU supervisors, such as the European Banking Authority and the European Securities and Markets Authority, in the upcoming months to contribute to making MiCA a success.

Asia

MAS to publish crypto regulation plans by mid-2023

MAS is now reviewing feedback on proposed crypto regulations. Response is expected by mid-2023.

MAS once highlighted support for "digital asset innovation", but opposition to cryptocurrency speculation. They are even considering adding frictions to retail access even they see potential in applications like trading, cross-border payment, settlement, and whatnots.

Singapore Plans New Guidance for Banks to Vet Crypto Clients

SG authorities are collaborating with lenders to establish uniform standards for vetting cryptocurrency clients. The central bank and police are providing assistance in fine-tuning the vetting approach for digital asset service providers. This initiative encompasses a range of digital assets, including stablecoins, NFTs, and transferable credits. Despite the standardized framework, banks will maintain their autonomy in deciding client acceptance based on their individual risk appetites.

Developing Web3 — Keeping Upright, Innovating and Moving Forward Steadily

And HK is striving to regain its former majesty. the initiative for HK towards digital economy is quite robust. Despite recent fluctuations in the virtual asset market, the HK government aims to push Web3 forward. Authorities will adopt a strategy that emphasizes both proper regulation and promoting development, ensuring financial security, investor protection, and AML. A licensing system for virtual asset service providers will be introduced in June, and $50M will be allocated to Cyberport to accelerate the development of Hong Kong's Web3 ecosystem.

Chinese Banks Court Crypto Firms in Hong Kong After Mainland Ban

China's state-owned banks are supporting crypto firms in HK, as the city aims to become a major digital asset center with Beijing's backing. This development offers an opportunity for HK banks to capitalize on the growing crypto industry, following recent banking failures in the US.

NFT White Paper: Japan’s NFT Strategy for the Web 3.0 Era

Japan is emerging as a welcoming hub for Web3 projects.

Japan's Liberal Democratic Party released a white paper outlining the country's NFT strategy for the Web 3.0 era (Just befor ETHTokyo).

It recommends developing a national strategy focused on NFTs and crypto assets, appointing a Web 3.0 minister, establishing a cross-ministry organization, and creating a "one-stop consultation desk" for private sector consultations.

The paper addresses topics like NFT business development, IP rights protection, user protection, blockchain ecosystem health, and safeguarding social and legal interests.

Ecosystem Whereabouts

Ethereum

[ETH] Shanghai Upgrade: The Hot Potato, Beacon Chain Withdrawal

Recommend this article for getting the hang of Shanghai Upgrade:

EIP-4895 is a significant Ethereum upgrade designed to enable beacon chain withdrawals with minimal impact on network security. It introduces gasless, partial, and full withdrawal options for validators. Although there may be increased selling pressure during the initial 4-6 days, it's expected that staking demand will rise after the upgrade. The successful implementation of EIP-4895 is a crucial step toward a complete Proof of Stake Ethereum, and it's essential for the community to actively observe and participate in the testnets to ensure a successful upgrade.

[EigenLayer] The Next Frontier in Crypto: EigenLayer’s Vision for a Lower Cost of Capital

And meaningful fundraising news for Ethereum eco — Eigenlayer.

EigenLayer presents a groundbreaking innovation in re-staking, which dramatically reduces the capital costs associated with bootstrapping new blockchain networks. By utilizing ETH already securing the Ethereum network to reinforce new networks, EigenLayer diminishes the required capital outlay to incentivize, attract, and retain a decentralized cohort of top-tier validators. Consequently, this fosters a more accessible environment for visionaries to launch novel networks.

The EigenLayer Stage 1 Testnet is now in progress.

Highly recommend the video below, as it provides an excellent guide to understanding the comprehensive structure of EigenLayer.

And this single, powerful image schematizes why do we need a re-staking collective layer for Ethereum.

[Linea] Public Testnet Launch

Public blockchains, wallets, and developer tools may not be fundamentally different in nature. The business of public blockchains revolves around recording programs in blockspace and attracting users, making economies of scale crucial. It seems plausible that ConsenSys, already offering a wallet and developer tools, could venture into the realm of public blockchains as well.

Polygon

[Avail] Polygon spins off project Avail and co-founder departs

Anurag is leaving with Avail spinoff.

Avail as a modular blockchain with a data availability layer is uniquely positioned to help developers overcome the challenges of monolithic blockchains but falls outside Polygon's L2 solutions. Thus, I offered to continue leading Avail as a separate, self-funded entity. — from Anurag's thread

Wonder how the topology would be differentiated betw Avail and Celestia after this movement. Further article comparing Ethereum, Celestia, EigenLayer, and Avail for DA layer — A comparison between DA layers

[Polygon zkEVM] Introducing Immutable zkEVM Powered by Polygon: the Home of Gaming in Web3 in Partnership with Immutable

Immutable zkEVM, an EVM-compatible ZK-rollup powered by Polygon, will launch in the coming months. I don't see that many seats available for the game chains since the number of big gaming projects are declining... wonder how other chains would differentiate themselves as well.

[Polygon Supernet] Nexon heads into NFTs with MapleStory Universe and Polygon deal

MapleStory Universe will be powered by Polygon Supernet. MapleStory N will include NFT-based items and characters, with DApps allowing NFT recognition and value beyond MapleStory N.

Think Nexon's reasoning and strategy with blockchain game quite resembles that of CCP Games. They are eager to build a long-lasting open ecosystem with extended in-game asset utilities and better standardization.

I believe only handful of blockchain games will survive, since it doesn't make sense for small games without community-driven works and robust narratives will be eventually sustainable for a longer timeframe. But these games will stand for a long time even if not successful in a short-term revenue perspective.

Gotchichain is set to launch, also powered by Polygon Supernets. It will be the go-to blockchain for Aavegotchi games and dApps, utilizing GHST as the gas token for all transactions. — From Introducing Gotchichain: Aavegotchi X Polygon Supernets

Cosmos

[dYdX] V4 Milestone 3 is Complete

dYdX Private Testnet is set to begin on March 28th, lasting 2-3 weeks. Fully-public testnet and mainnet launch are planned for July and September 2023, respectively. Full node, validators, indexers, and frontends will be open-sourced.

dYdX is moving towards full decentralization in response to increased scrutiny from global regulators. The exchange is launching a test network for the next version of dYdX, which aims to remove centralized entities currently maintaining its operations. Founder and CEO Antonio Juliano believes that full decentralization is the ultimate goal for DeFi projects. — From DeFi Exchange DYdX Seeks Full Decentralization as Scrutiny Rises

[Noble] Announcing USDC on Noble

Noble is set to revolutionize the IBC ecosystems with native USDC.

  • This integration aims to boost liquidity in Cosmos and address the challenges with bridged assets.

  • Noble has been developed through a collab betw Informal Systems, Strangelove Labs, and Iqlusion, with Strangelove as the primary engineering partner.

  • The platform aims to simplify asset ownership and transfer within the IBC ecosystem, starting with USDC and eventually supporting a wide range of digital assets.

And the dYdX chain will use native USDC powered by Noble at launch.

For more info: Noble_xyz is more than "USDC on Cosmos"!

[Injective] Injective Launches Cascade: First Ever Interchain Solana SVM Rollup

  • Injective is launching Cascade, the first-ever Solana SVM RU for IBC, enabling Solana devs to deploy their contracts and dApps onto Injective.

  • The Solana Sealevel Virtual Machine (SVM) is a parallel smart contract runtime, while Eclipse is a customizable RU provider.

  • Cascade's launch will go through several deployment phases, with the goal of having a permissionless sequencer and Injective as the core settlement layer.

Tokenization

Citi Report

Citi Says Mass Adoption of Crypto Will Be Driven by CBDCs, Tokenization

Citi analysts expect $5T circulating in CBDCs by the end of the decade, with some having blockchain interoperability or being DLT-specific. The advantages include interoperable payment instruments and enthusiasm from developing economies. However, risks include protecting user privacy and users shifting deposits from smaller commercial banks to CBDCs.

Tokenization of traditional financial assets could be the "killer use case" for blockchain, potentially growing 80x in private markets and reaching nearly $4T by 2030. Benefits include disintermediation, composability with cryptocurrencies, and a shared "golden-source" infrastructure for different asset classes. Challenges to overcome include regulatory clarity and potential pushback from the financial industry, as tokenization could render some jobs obsolete.

The envisioned end state is a digitally native infrastructure that operates 24/7 and is enhanced with smart contract and DLT-enabled automation. This infrastructure would enable new product features, improve market access, and streamline distribution. Despite skepticism among some traditional players, the collective value of tokenization's benefits has the potential to drive significant transformation in the financial sector.

Leading ecosystem players are bullish on tokenization due to its potential benefits. Traditional financial market infrastructures (FMIs) are the primary focus, as they possess a large user base, trade volumes, and assets under management. Early movers could gain a significant advantage in a market worth trillions. Tokenization can drive new portfolio opportunities, change market structures, and reward entities that adapt to the new model.

Always thought that how we assess the cryptocurrency market cap in the future would be somehow different from now bcz I believe securities, especially private market assets; private equity & real estate, et al will eventually be tokenized.

It's already happening:

Come for the Tool, Stay for the Exchange

Come for the Tool, Stay for the Exchange: Bootstrapping Liquidity in the Private Markets

Read this last year while studying for Carta; I'd share it again.

The shift from public to private markets has increased private market assets to over $6T globally. To meet the growing need for liquidity, a new generation of private market exchanges is expected to emerge, beginning as software tools.

The "Come for the Tool, Stay for the Exchange" strategy can help entrepreneurs bootstrap liquidity, avoid negative selection, and ensure common financial reporting standards.

Companies like Carta and Juniper Square serve as examples of businesses executing this strategy, creating two-sided networks and unlocking significant enterprise value by creating durable venues for private market liquidity across various asset classes.

$1 trillion in equity: How Carta is set to unlock the private markets

Carta aims to address this by launching CartaX, the first issuer-sanctioned platform for buying and selling private company shares at scale. Carta's dominant position and strong network effect allows it to create a trusted environment for controlled auctions, enabling private companies to effortlessly run auctions while maintaining the benefits of staying private.

CartaX is expected to be a game changer for companies, employees, and investors, providing better attraction, incentivization, and retention of talent and capital. It will help close the liquidity gap in private markets, which currently have trading volumes 330 times lower than public markets. As CartaX becomes a critical component for startup success, the venture ecosystem may experience a significant behavioral shift.

AFAIK, there are at least two companies building 'Carta for crypto' including Magna and LiquiFi.

Magna raises $15m at $70m round led by Tiger Global and Tusk Venture Partners

LiquiFi selected for YC W22

Not really happened so far, but likely to happen in the near future:

I think that all new issuances will be tokenized five years from now. I wouldn’t be surprised if that happens more quickly. Perhaps in three years, every new stock, bond, anything being sold…is actually a token. But there would still be the legacy instruments out there — probably ten years from now it will all have been rolled over. So, I think that’s probably the correct way to say it. Within three to five years, all new instruments will be coming out as tokens, and by ten years all instruments will have rolled over to be tokens. I would imagine that’s the truth. — From Patrick Byrne Discusses Security Token Future, tZERO Updates, Ravencoin, More

Investments

Fintech Valuation

eToro secures $250M at a $3.5B valuation after scrapping SPAC, seeing slower growth

Stripe down 45%, Klarna down 85%, FTX DOWN BAD 99.9%, and now eToro down 65%. Revolut the only surviver among mentioned above (because they didn't get the fresh injection yet)?

eToro secured $250M at $3.5B, following the cancellation of its SPAC at $10.4B last year. The funding comes via an AIA from ION Group, Social Leverage, SoftBank, and Spark Capital. The company reported 2.8M funded accounts by the end of '22, up from 2.4M in '21. Commissions in '22 totaled $631M, down half compared to '21 and only up 5% from '20.

One thing on my mind is that fintech is a byproduct of a cyclical phenomenon. We used to play around with money to make money at the end of each bull cycle, and then some of the noted ones would go bust. This would result in a reset and a retreat of regulatory flexibility.

Eventually, a recession would come, and another cycle would begin. I feel like we are currently at the edge of a cycle, which means that another set of hope and opportunities is here, yet to rise.

Saudi’s Investment

The Saudis are investing. A situation where there is an urgent need to turn resources into national infrastructure before they run out. Spending money well is just as difficult as making money well. But these days, as I watch the Saudis tiptoe between the US and China, I think they're doing a pretty good job.

Saudi Arabia Discloses Ties to Andreessen Horowitz, Dozens of Other Venture Funds

PIF has revealed its venture arm, Sanabil Investments (investing $2B annualy in VC, growth, and buyout funds), has ties with around 40 US-based VC firms, including a16z, Coatue, and Insight Partners (and some of the noted crypto funds including Polychain, Iconiq, and Race) through website.

The relationship between Silicon Valley and Saudi Arabia has been controversial since the 2018 murder of journalist Jamal Khashoggi, but the sovereign wealth fund has managed to distance itself from the country's politics through a public relations campaign.

Saudi Arabia Is Investing $38 Billion to Become a Video-Game Hub

Think I've seen this before like couple months ago, but uploaded again on major media again. Saudi's initiative towards gaming is aggresive. (But some critics have described Savvy's esports initiatives as "sportswashing.")

Saudi Arabia is investing $38B through PIF to become a major global gaming market player. Savvy Games Group, the PIF's subsidiary, aims to develop, publish, and acquire top-tier games and support the gaming industry in Riyadh.

In 2022, Savvy acquired ESL, an esports tournament company, and merged it with gaming platform FACEIT in a combined $1.5B deal. In March, EFG acquired esports technology company Vindex. — ESL FACEIT Group Acquires Vindex to Strengthen Position as Global Leader in Esports & Video Game Entertainment

One more recent deal to mention: Saudi’s Savvy Games Group to acquire mobile games company Scopely for $4.9 billion

... and Some Others

2022 Annual Letter by Chamath Palihapitiya

Social Capital's Annual Letter. Chamath brings two core themes driving future innovation: energy and compute costs approaching zero.

Theory

And Tomasz Tunguz, ex-Managing Partner at Redpoint, announced his new venture firm, Theory, with launch of $230M fund. Think Theory Ventures can be classified as 'Data VC.'

The 3 main areas he's targeting for:

  1. Decade of Data (data software for modern organizations)

  2. Machine Learning as Force Multiplier (enhancing workflows and user needs)

  3. Decentralized Infrastructure as Database (blockchain transforming user and asset relationships)

Crypto VCs Revert to What’s Safe: Blockchain’s Building Blocks

Crypto infrastructure startups are attracting investors who see them as a safer option. This investment trend mirrors the post-dot-com crash focus on internet back-end software companies.

Crypto infrastructure firms also pose lower regulatory risks compared to token-issuing businesses. The growing number of active blockchain developers further supports the infrastructure sector's appeal, as these firms primarily serve other cryptocurrency businesses.

Well, that wraps up the first edition of Facta non verba. I might've bitten off more than I can chew with the length of this week's content, so I can't promise I'll keep it up for the next one. But if you found it useful, please hit that subscribe button!

And as a little incentive, the first five subscribers can collect this article for just the price of a cup of coffee (0.002 ETH). See you next time!

]]> danpark@newsletter.paragraph.com (Dan Park) ai crypto investment