Let me start somewhere that feels distant but is not.

In the 1960s, if you wanted to understand the future of computing, you studied IBM. Mainframes. Centralized infrastructure. Enormous capital requirements. A small number of institutions controlling access to computation on behalf of everyone else. The assumption was simple: scale implies concentration. Concentration implies control. Control implies permanence.

Then something happened that looked, at first, like a toy.

Personal computers were worse than mainframes on almost every measurable metric. Slower. Less reliable. Fragmented. Inefficient. IBM itself launched one largely as a defensive move, not expecting it to matter. But the PC did not win by being a better mainframe. It won by introducing a completely different architecture: distributed ownership, modular software ecosystems, open participation, local autonomy. The bottleneck shifted. It moved from compute production to compute deployment. From centralized control to distributed participation.

IBM did not dominate the personal computing era. Not because they lacked capital or talent but because they were optimizing the wrong layer of the stack.

The railway winner did not build the automobile system

Go back further and the pattern repeats.

The great railway companies of the nineteenth century were the hyperscalers of their time. Enormous capital concentration. Critical infrastructure. National scale. And they were genuinely transformative — for their era. But the winners of rail did not automatically become the winners of the automobile age.

Why not?

Because the shift from rail to automobile was not merely technological. It was architectural. The automobile required roads, not tracks. Distributed fuel distribution, not centralized stations. Individual ownership, not ticketed access. Suburbs, not rail hubs. The entire coordination model changed. A new system had to be built around the new technology, and the incumbents who had optimized for the previous system were often the last to see it. The decisive factor was not the combustion engine. It was the governance and economic architecture that emerged around it.

Europe had the Internet before the Internet

There is a version of this story that is particularly relevant for Europeans.

France had Minitel. A sophisticated, state-backed network, operational from the early 1980s, providing services that the rest of the world would not see for another decade: online shopping, train reservations, messaging, information services. Technically impressive. Well-governed. Nationally deployed.

The open Internet ate it alive.

Not because TCP/IP was technically superior in every respect. But because the Internet's architecture was fundamentally different: permissionless. Composable. Open. Anyone could build on it without asking permission. Anyone could participate in creating value, not just consuming it. The scaling logic was distributed participation and not centralized provision. Minitel optimized one paradigm brilliantly. The Internet invented a different paradigm entirely.

Europe, which spent years mourning this loss as a cautionary tale about missing the wave, may now be misreading its own history. The lesson is not: move faster on the current paradigm. The lesson is: recognize when the paradigm itself is about to change.

Intelligence is becoming a commodity

Here is what I think is actually happening right now.

Open-source and open-weight models are compressing the capability gap between frontier proprietary systems and widely accessible models faster than almost anyone predicted. The cost of inference is falling. The difference between what you can run locally and what requires a hyperscaler data center is shrinking every quarter — not at the margins, but at the core. Intelligence is becoming a commodity, so the serious question is no longer whether it will happen, but what comes after.

And this is exactly where the interesting thing occurs: Once intelligence becomes abundant, the bottleneck shifts. It moves from producing intelligence to coordinating it. From centralized ownership of models to distributed deployment of capability, and from who has the largest model to who has the best architecture for putting intelligence to work at scale, across many actors, in ways that generate real productive value. So, the decisive variables become deployment, orchestration, legitimacy, trust, resilience, and participation in value creation. In other words: governance.

Not in the sense of regulation imposed from outside but in the sense of production infrastructure. Embedded, architectural, native to the system.

The Renaissance was not just about art

This is where history becomes genuinely useful rather than merely decorative.

We tend to remember the Renaissance through its most visible outputs: painting, sculpture, architecture, science. But the Renaissance was equally a revolution in coordination architecture. Double-entry bookkeeping. Merchant banking. Commercial insurance. The letter of credit. Republican civic institutions. The university. The printing press as distribution infrastructure.

All of these where not only economic, artistic or scientific inventions but also institutional inventions. New ways of coordinating economic activity, sharing knowledge, organizing production, distributing surplus, governing collective decisions. The artistic and scientific flowering was possible because the coordination infrastructure underneath it changed.

The industrial revolution followed the same pattern. The steam engine was real, but it was not sufficient. What transformed society was the railway as coordination infrastructure, industrial finance, standardized time zones, mass public education, corporate law, logistics networks, the factory as organizational form. The economic transformation required an institutional transformation.

I think AI is now at a similar inflection point.

The models exist. The capability exists. What does not yet exist — what has not yet been invented — is the coordination architecture that will allow that capability to generate its full civilizational impact. That architecture is the next frontier.

Chasing the hyperscaler is the Minitel move

This is the uncomfortable implication for anyone thinking about Europe's AI strategy.

The current instinct — sovereign compute, national champions, catching up on model scale — is an inversed version of the Minitel move. It tries to optimizes the previous paradigm. It competes on the incumbent's terms. And it arrives late, with less capital, in a race where the structural advantages have already been consolidated elsewhere. Consider a minute that US top companies plan to spend almost 2.000.000.000.000$ of Capex for AI in 2026: a million of millions of millions (do I even get this right?).

Even leaders of the most advanced AI labs increasingly acknowledge a version of this problem. If AI generates enormous productive abundance while simultaneously disconnecting most people from participation in that production, redistribution becomes the only remaining policy option, which is why UBI increasingly enters the conversation. But redistribution after production is not the only possible architecture.

The deeper question — the actually frontier question — is whether participation in AI-driven production can itself be distributed. Whether the coordination layer of AI can be designed so that more people are participants in value creation, not merely recipients of whatever the hyperscalers choose to share.

That is not a welfare question. It is an architecture question.

The two paths

I find myself back where I ended after the displacement of scarcity experiments (see there: https://paragraph.com/@antoinevergne/the-displacement-of-scarcity.

There are two genuinely possible paths.

The first is hyperscaler consolidation. A small number of actors control the models, the compute, the deployment infrastructure, and the economic surplus generated by abundant intelligence. Redistribution mechanisms — UBI, wealth taxes, public funds — attempt to route some of that surplus back to people who have been structurally disconnected from production. It is not a dystopia. It may work, partially. But it is a world where most people are passengers in an economy they do not participate in building.

The second path requires more imagination. It starts from the same observation — intelligence is commoditizing, the bottleneck is shifting to coordination — and asks: what if the coordination architecture itself were designed for distributed participation? What if governance were not added on top of AI systems as an afterthought, but embedded into their production architecture from the beginning? What if the next Frontier AI Lab were not a hyperscaler at all, but a coordination-native institution — one that scales through distributed participation rather than centralized ownership?

This is what I have been calling Syntropolis. Not a governance project layered over AI. A different theory of what the productive architecture of AI could look like once intelligence stops being the scarce resource.

The question is no longer whether AI will create abundance

The question is whether participation in AI-driven production can scale as fast as intelligence itself. That is the actual frontier hypothesis.

And if history is any guide — from railways to automobiles, from mainframes to personal computing, from Minitel to the open Internet, from craft production to industrial capitalism — the answer will not come from optimizing the current paradigm more efficiently, but it will come from someone who noticed that the scaling logic had changed.

The future of AI may not belong to the largest model, it may belong to the best coordination architecture.

This essay builds on The displacement of scarcity and is part of an ongoing exploration of Syntropolis — a coordination-native model for the AI economy.

An API — Application Programming Interface — is, at its core, a formal agreement between two systems about how they will talk to each other. You ask. It responds. Everyone pretends this is straightforward. In this way, the API is not unlike marriage.

For most of its existence, the API lived a quiet and largely unloved life, deep in the documentation that nobody read, maintained by people whose idea of a good Friday night was arguing about the correct way to name a subroutine. It was, by any reasonable measure, extraordinarily boring. It was the beige wallpaper of technology. It was the terms and conditions of infrastructure. It was the thing that made eyes glaze over at dinner parties, which is why software engineers stopped mentioning it at dinner parties, which is why software engineers stopped getting invited to dinner parties.

There was, however, at least one exception.

In 1983, a French software developer named Gérard Plombier, working nights at a telecommunications firm in Lyon on a project that was already eighteen months late and spiritually abandoned, wrote the following sonnet in the margin of an API specification document. The API in question governed the exchange of data between two mainframes, each the size of a wardrobe, neither of which was connected to anything resembling the internet, because the internet did not yet exist in any form that would have recognised itself in the mirror. It was, by the standards of things nobody cared about, something that even fewer people cared about.

Gérard cared.

Ô douce Interface, promesse sans visage, Tu reçois ma requête avec grâce infinie, Et dans ton protocole, ô merveille benie, Tu me rends une réponse — parfois — sans dommage.

Nul homme ne te voit, nul poète ne t'engage, Tu œuvres dans le calme, ignorée, incomprise, Pendant que le client crie et que la deadlinerise, Toi seule réponds « 200 OK » au passage.

Un jour viendra, je le sens dans mes octets, Où les machines entre elles tisseront leurs secrets, Et c'est toi, humble interface, qui tiendras les fils.

On ne te remerciera pas. On ne te comprendra pas. Mais le monde tournera grâce à toi, tout bas, Ô toi, ma bien-aimée. Ô toi, mon API.

— Gérard Plombier, Lyon, 1983 (Traduit de l'obscurité par l'auteur)

 sweet Interface, promise without a face,
You receive my request with infinite grace,
And in your protocol, oh wonder most blessed,
You return a response — sometimes — unaddressed.
No man sees you, no poet seeks your embrace,
You toil in the quiet, ignored and misread,
While the client shouts and the deadline fills us with dread,
You alone reply "200 OK" and hold your place.
A day will come, I feel it in my bytes,
Where machines shall weave their secrets, one to one,
And it is you, humble interface, who holds the threads.
No one will thank you. No one will understand.
But the world will turn, softly, by your hand,
O you, my beloved. O you, my API.
— Gérard Plombier, Lyon, 1983
(Translated from the obscure by the author)

The poem was never published. The project was eventually cancelled. Gérard has since pivoted to artisanal cheese, in which he has found, he says, greater reliability and more predictable response times.

And yet Gérard, writing in the margins of a document that governed the conversation between two wardrobes in Lyon, had seen something that the rest of the world would take forty years to notice.

AI arrived.

Not with a bang, exactly. More with a very confident, occasionally hallucinatory, and surprisingly articulate hum. Suddenly there were models that could write your emails, summarise your meetings, book your holidays, and explain quantum mechanics in the style of a disappointed Victorian uncle. The world was impressed. The world was slightly unnerved. The world, by and large, did not ask how any of it worked.

The answer, inevitably, was APIs.

Because here is the thing about intelligence, artificial or otherwise: thinking is one thing, but doing requires talking to other. When an AI books your dentist appointment, it calls an API. When it reads your emails, it calls an API. When it confidently summarises a PDF it has somehow not actually read, it calls an API, and then probably several more to cover its tracks.

And now — and this is where it gets genuinely strange — the AIs are beginning to talk to each other. Agents are spawning sub-agents. Models are calling models. Digital assistants are hiring digital assistants. An instruction issued by a human at breakfast may, by lunchtime, have passed through seventeen different artificial minds, none of whom have had coffee, all of whom are extremely responsive and available at all hours, and every single handoff between them happens through an API.

The beige wallpaper is now load-bearing.

The humble API — long the exclusive obsession of people who named their children after sorting algorithms — has become something rather more significant than anyone planned. Every API is a boundary. A gate. A decision, encoded quietly in a schema, about what is permitted, what is exposed, and what one intelligence is allowed to ask of another. They are not technical documents anymore but are becoming - without anyone having called a vote, organised a summit, or a global protest march - the constitutional framework of a civilisation run by agents.

The nerds have been writing the rules all along. They just used very small fonts, and nobody checked.

Gérard knew.

HTML2, Counter strike, Web3

I have always been fascinated by technology. It started with a commodore 64 in 1989, then with an Olivetti desktop computer in 1995. A year later, I managed to convince the director of my high school to get a subscription to access the Internet to be able to code the webpage of our high school. It was in HTML2, in a text editor, with many animated GIFs. The 2000s and the era of broadband, online gaming, was also a thrill. I missed web2 and social media, got very interested in virtual worlds around 2015, then Web3 and crypto took over my attention. So much potential, so many pitfalls. Synthetic intelligence has been also a core topic since 2020 and the results of “We, the internet” the global dialogue on the future of the Internet during which citizens of the world gave one core advice: “we need to talk about AI much more, have thousands of dialogues”. How right they were 😊

Last November I entered a new rabbit hole: Vibe coding. Synthetic intelligence had been part of my curiosity, I followed the field, read about it, experimented as soon as an LLM came out. I could see the potential, but it remained somewhat abstract. I had ideas for applications, pieces of software, even business models. They were intuitions more than projects.

The obstacle was simple: I did not know how to code well enough to turn those ideas into real products. HTML 2 was not that up-to-date, hiring a team was unrealistic. Raising capital made no sense. There was always a gap between imagination and execution.

The vibe coding rabbit hole and the decentralized stack

With vibe coding, that gap started to close.

Things that would have taken weeks suddenly became possible in a few hours. Projects that would normally require a team suddenly felt accessible to a single person. The technical barrier did not disappear entirely, but it stopped being decisive.

For the first time, I could test my ideas directly. I coded a swarm of agent to help me write an article; I deployed a custom smart home dashboard; I tested a couple of autonomous trading agents on Web3 platforms; deployed a full website in 4 hours with content at the right place, a nice design and in two languages. During a weekend, I recreated a civic tech platform that had taken us 3 years, 6 devs, 10 partners and €3M in an EU project to create.

Very quickly, a second step followed. If I could build things this way, why depend on proprietary tools? I started using open-source software for my projects. Then I moved to open-weight models and an open-source coding assistant.

What I discovered was both exciting and slightly uncomfortable: these models were good enough. Not perfect, but more than capable of building real things, iterating. Good enough for perhaps ninety percent of what anyone needs to do.

The next step was logical: If this worked, why stay dependent on cloud infrastructure? I used my Dappnode (small server the size of a chocolate box) to deploy the models locally, run the agents locally, store all data locally. For sure, I learned to deal with hardware limitations, to optimize, to wait a bit longer for answers, to work within constraints as I have no GPU or high-end hardware. But still, it was 80% as good as the cloud frontier model.

This lead me to the next step: if I was talking about decentralization, why remain dependent on centralized energy systems?

No reason, so I plugged my stack into my balcony solar powerplant and a home battery. For sure, its an investment, but one that is amortized in a couple of years. And it can also power my dishwasher and my oven.

Abundance?

So over the course of 2 months, I realized that the entire stack — open-source software, open-weight models, a local server, partially autonomous energy — could run with extremely low marginal costs once the initial infrastructure was in place. That was the moment when I felt something that looked like abundance.

Not universal or material abundance, but an abundance of usable intelligence and capacity to build software: I could create almost anything I could imagine, without asking anyone for permission or having to discard the idea right away.

And yet, nothing else around me had become more abundant. Food was not free, it was even rising quickly due to inflation, same for housing. Land was not abundant and physical infrastructure still obeyed the old logic of scarcity and power. My balcony power plant was still worth a year of salary in some countries of the world, the server too.

The economics do not match

This opened a new set of questions. If I, individually, could produce software at almost zero cost, what happens when millions of people can do the same thing? The answer is not abstract. The SaaS model — the dominant business model of the past two decades — rests on a specific friction which is that building software is hard, that distribution requires infrastructure, that switching costs create lock-in. Remove those frictions, and the model collapses. We have seen that in the past weeks with each new skill from major AI models.

Why pay a monthly subscription for a tool that any moderately skilled person can now replicate in an afternoon for quasi free? And what happens down the line in 5 to 10 years when capability has risen, and an agent guiding other agents will create the software you need on the go on your phone?

The hyperscalers face a version of the same problem, only more structural. They built their moats on proprietary models, on the assumption that frontier intelligence would remain expensive and centralized. But open source, open-weight models are closing the capability gap faster than almost anyone expected. The cost of inference is falling. The difference between what you can run locally and what requires a data center is shrinking every quarter — not at the margins, but at the core. This is not a temporary lag. It is a fundamental erosion of the business model.

The largest capital investments in history are being made in AI infrastructure — chips, data centers, energy grids. Those investments require a return which requires scarcity: the ability to charge for access to something others cannot replicate. But the technology itself tends toward abundance. Every open release, every efficiency gain, every model that runs on consumer hardware undermines the scarcity the infrastructure is built to monetize. Part of it is the consequence of the arms race between the US and China which goes full steam open source.  

The interesting part is that it is not only software where this logic applies. Current AI, a public interest initiative born at the Paris AI Summit, recently unveiled a handheld device built with the Indian government that runs three simultaneous AI models locally, across twenty-two languages, with no internet connection, on open-source hardware anyone can replicate. That is not a research prototype. So, even the hardware layer — the last plausible bottleneck for centralized control — is being opened.

Conclusion: You cannot build a durable rent-extracting monopoly on a commodity.

The displacement of value

So where does value actually come from and goes? This is the central economic question of the next decade. Some analysts have tried to answer it through financial scenario analysis — tracing how SaaS collapses into private credit defaults into mortgage stress into a broader demand spiral. That analysis is serious and the transmission mechanisms are real. But it frames the wrong crisis. The financial contagion is a consequence. The structural fact underneath it is simpler and more disorienting: intelligence is becoming a commons, and the institutions built to monetize its scarcity have no obvious successor. What I experienced was not the end of scarcity. It was the displacement of scarcity. Intelligence becomes cheap. Coordination becomes valuable. Access to physical resources remains structured by institutions and power.

As a side quest: The regions that invested most heavily in the first era of frontier AI — that raced to build proprietary models, sovereign data centers, national champions — are now most exposed to the model becoming obsolete. The EU, which spent years criticizing itself for missing the first wave, may have stumbled into an advantage: it arrives at the next era without a legacy infrastructure to protect, without national champions whose business models depend on keeping intelligence scarce. If the next era belongs to open, distributed, public-interest AI, then not having bet everything on the previous era is not a handicap but optionality.

More questions

From there, the questions that stayed with me were not about which companies survive but about what happens to people. If we can produce more with less human labor, why should survival remain tied to employment? And if work stops being the primary source of income, what happens to status? What happens to recognition? And what happens when synthetic intelligence becomes embodied and replaces millions of people?

Large societies that have partially separated survival from work did so deliberately and after long struggles: Public pensions decoupled old age from labor, universal healthcare decoupled illness from income, and public education decoupled learning from wealth. None of that happened automatically because technology made it possible but because institutions were redesigned after political struggle that took decades.

If intelligence truly becomes cheap, we are facing a similar moment. But this time the thing becoming cheap is not a particular category of work — it is the cognitive substrate of almost all work. The displacement is general which is something that many observers seem to downplay when they compare the current revolution to past technological leaps.

The path forward

Based on these little experiments in my living room, I see two paths that are genuinely possible.

The first is ungoverned abundance. The open-weight ecosystem wins. Intelligence becomes a genuine commons. Individuals and small groups gain real capacity. But the collective action problems — inequality, governance, climate, coordination — remain unsolved, because no institution captures the surplus generated by cheap intelligence and routes it toward solving shared problems. In this world, people have extraordinary tools and literally no roof. Digital abundance coexists with material insecurity. Extraordinary software explains you why you cannot afford rent or why butter is twice as expensive as last year.

The second path requires more imagination, and more work. It starts from the same observation — that intelligence is becoming a commons — and asks what kind of institution could govern a commons at that scale. Not a state in the traditional sense, and not a market. Something that treats the surplus generated by cheap intelligence as shared infrastructure, and organizes its governance around participation rather than ownership. I have been thinking about this under the name Syntropolis https://paragraph.com/@antoinevergne/syntropolis-a-blue-print-for-mastering-ai : a polity built not on territory but on shared stewardship of synthetic intelligence, governed through randomly selected citizens' assemblies, distributing its surplus as a minimal guarantee of dignity: If intelligence is becoming the primary productive infrastructure of the economy, then inclusive control of that infrastructure is the condition for anything else being possible.

I do not know which path we will take.

What I do know is that the choice is not technical. The technology works. Open models exist. Local compute is real. Hardware is opening. The economics of abundance are not speculation and the question is whether we are willing to organize what comes next — or whether we will let it organize us. Abundance has to be built.

Money, money, money

Here is something strange about money that most people never think about. When you ask what a euro is worth, the answer is: a euro is worth a euro. The European Central Bank does not tell you that one euro equals three days of groceries, or twelve kilowatts of electricity, or four months of internet access. It tells you, essentially, that one euro is worth the things that one euro can currently buy — which, of course, changes every year, every month, sometimes every week.

This is not a trivial problem. It is the founding instability of every currency on earth. Money is classically supposed to preserve value across time (be a store of value). But if the unit of measurement is itself floating — if 'a dollar' means something different in 2005 than it does in 2025 — then money isn't actually preserving anything. It's a measuring tape that slowly shrinks.

For most people, this shows up as a acceptable inconvenience. Savings erode a little. Salaries lag behind costs. You notice that the grocery bill is somehow higher even though you're buying the same things. But for a significant portion of humanity — people who spend 60, 70, sometimes 80 percent of their income on food, energy, and shelter — this instability isn't an inconvenience. It's a structural catastrophe that money inflicts on them slowly and invisibly, year after year.

This essay is about a proposal to fix that. Not by tweaking interest rates, or redesigning central banks, or creating another cryptocurrency. By rethinking what the unit of money actually means. By building a currency that doesn't say 'I'm worth a euro.' It says: 'I'm worth a month of decent human life.'

Most currencies promise to be worth something. SAWA promises to be worth something specific — the things a human being actually needs.

The invisible tax

Let's start with something concrete. In 2005, a working-class family in Lagos could feed four people for a month on roughly ₦8,000. By 2023, the same basket of food cost ₦80,000 — a tenfold increase over eighteen years (Source: World Bank CPI for Nigeria). The Nigerian naira, officially, maintained its existence. The Central Bank published reports. The government collected taxes. The exchange rate moved, as exchange rates do. But in real terms — in the terms that matter to a mother buying rice — the currency had been quietly robbed of most of its meaning.

This is not a story unique to Nigeria. In Argentina, Turkey, Lebanon, Zimbabwe, Venezuela, and dozens of other countries, inflation has operated as a silent expropriation — a tax levied by no one in particular and paid disproportionately by the people who can least afford it. But even in stable economies — Germany, France, the United States — essential goods tend to inflate faster than composite price indices, meaning that even 'low inflation' environments concentrate purchasing power erosion at the bottom of the income distribution.

The mechanism is not malicious. It's structural. When central banks create money through credit expansion, the new money flows first to financial actors — banks, corporations, asset holders — who use it to buy assets before prices adjust. By the time the purchasing power reaches wage earners in the form of salaries and transfers, prices have already risen. This isn't a conspiracy. It's how the plumbing of money works. The people at the end of the pipeline always get the least.

There is a technical term for this: the Cantillon effect. Named after an eighteenth-century economist who noticed that who receives new money first determines who benefits from its creation. The modern financial system has elevated the Cantillon effect into an institution. Central banks create money. It flows to banks. Banks lend to corporations and asset markets. Asset prices rise. People who own assets become wealthier in nominal terms. People who own wages and savings — most people — find themselves slightly poorer in real terms, year after year, without anything obviously going wrong.

What If Money Was Measured in Real Things?

The obvious question is: what if you simply measured money in terms of the things people actually need? Not in terms of itself — not 'a euro is worth a euro' — but in terms of a specific, verifiable, end-user determined basket of essential goods?

This idea is not new of course.

In 1913, the American economist Irving Fisher proposed the 'compensated dollar' — a currency whose gold content would be periodically adjusted to maintain stable purchasing power against a commodity basket. The proposal was elegant in theory and politically dead on arrival. The required adjustments would be slow, centralised, and discretionary. More fundamentally, Fisher's system had no redemption mechanism: there was no way for an ordinary person to actually exchange their dollars for the basket. Without enforcement, a peg is just a wish.

In 1943, John Maynard Keynes proposed the Bancor — an international currency for settling trade between nations, based on a basket of commodities. The Bancor never happened. The United States, emerging from World War II as the world's dominant creditor, preferred a system anchored to the dollar. The International Monetary Fund's Special Drawing Rights are a distant descendant of the Bancor: a basket-weighted synthetic currency used only between central banks, inaccessible to any individual person.

Chile's Unidad de Fomento — the UF — has actually worked since 1967, indexing long-term contracts to inflation so that mortgage payments and savings accounts maintain real value over time. It is the closest real-world precedent to SAWA. But the UF is not a currency you can hold. It's a unit of account that settles in pesos at an inflation-adjusted rate. When Chile experienced currency crises, UF holders still got pesos — just more of them. Physical goods were not guaranteed.

The pattern across all prior attempts is consistent. They either stayed abstract (inter-governmental units that never touched ordinary people), lacked enforcement (pegs with no redemption mechanism), or remained passive (indices that adjust nominal values without guaranteeing real access). None of them gave a person on the street the ability to walk into a shop and exchange their savings for the actual things those savings were supposed to represent.

Previous attempt at basket-backed money failed for the same reason: it made a promise it couldn't keep. SAWA aims at tackling this.

The Three Innovations SAWA Combines

SAWA — the name comes from the Swahili and Amharic word meaning 'fine,' 'adequate,' 'just right' — is a monetary architecture designed by synthesising lessons from all these failures. Its proposal is simple to state, though operationally complex: define the unit of account as a redeemable claim on a specific basket of essential goods, governed by democratically elected citizens’ assemblies, with a technical architecture that makes the guarantee physically enforceable.

Three things make SAWA genuinely different from other proposals I found until date:

First: Redeemability that actually works

In SAWA, redeemable doesn’t mean “we'll give you a different kind of financial instrument.” It means you can walk into a partner supermarket, pharmacy, or utility provider and exchange your tokens for specific quantities of food, energy, hygiene products, and mobile connectivity. Goods and services. Not a promise denominated in another currency. The actual things.

This isn't stored in a warehouse somewhere. SAWA operates on the same Just-In-Time model that already runs modern supply chains: supermarkets replenish daily, electricity grids produce on demand, mobile data is allocated digitally. SAWA redemption flows through existing supply infrastructure rather than requiring dedicated inventory. Partner merchants — supermarkets, pharmacies, energy providers, telecoms — provide goods on demand in exchange for SAWA tokens, earning guaranteed revenues and more predictable demand.

The economic logic is self-correcting. If SAWA tokens trade above their basket value, holders redeem them for goods and the price comes back down. If they trade below basket value, buyers purchase more and the price rises. No central bank intervention required. No policy committee required. The peg enforces itself because redemption creates constant arbitrage pressure. This is one of the mechanisms prior proposals seem to lack.

Second: Governance by citizens

Who decides what goes in the basket? This is not a trivial question. 'Adequate nutrition' in northern Norway looks nothing like adequate nutrition in coastal Senegal. 'Basic mobility' in a dense European city means a transit pass; in rural Mali it means something else entirely. Any system that imposes a single global basket is either meaninglessly abstract or culturally coercive.

SAWA delegates basket composition to Citizens' Assemblies — randomly selected groups of ordinary people, stratified to be demographically representative of their community — organised at four levels: local, national, regional, and global. Assembly members don't need any special expertise. Their job is to deliberate and decide what constitutes a decent life in their context. Independent experts provide testimony. Deliberations are public. Decisions are binding.

The mechanism for selecting assembly members is called sortition — essentially democratic lottery, weighted for demographic balance. You've seen it in action without knowing it: the Irish Citizens' Assembly that broke the country's abortion deadlock, the French Convention Citoyenne pour le Climat, the citizens' assemblies that have navigated contentious issues in Iceland, Canada, and Belgium, the 12+ European Citizens’ Panels that tackled topics like Virtual Worlds, the EU 2 trillion budget, Intergenerational Fairness, etc. Sortition produces decisions that are representative, informed, and durable.

Applied to monetary policy, this is genuinely radical. Central banks are among the most deliberately insulated institutions in modern democracies. The idea that ordinary people — a Lagos market vendor, a Lyon bus driver, a Mumbai schoolteacher — should have binding authority over what their currency represents is not how monetary policy has ever worked. It is, however, how it probably should.

Third: Two layers of tokens — one for dignity, one for freedom

SAWA has two distinct token types.

pSAWA — soulbound provisioning tokens. These are non-transferable, non-sellable digital tokens tied to your identity. They represent your Universal Basic Provisioning: a guaranteed monthly floor of nutrition, energy, connectivity, and hygiene. You cannot sell them. You cannot be coerced into selling them. You can only redeem them for the goods they represent, at which point they are destroyed. They exist purely as a capability guarantee — a technically enforced floor below which your real living standard cannot be pushed.

mSAWA — transferable market tokens. These are ordinary tokens you can hold, save, transfer, earn as salary, send across borders, invest. They represent SAWA's function as a currency — stable, real-value-anchored, but free to flow wherever economic life requires. A Senegalese worker in Paris can earn mSAWA in euros and send them to his family in Dakar, where they're worth the same amount of food regardless of what happens to the CFA franc.

Why the separation? Because the history of tradeable welfare instruments is grim. Food stamps, ration cards, and welfare vouchers have been systematically extracted from vulnerable recipients through economic pressure, loan-shark schemes, and outright coercion. If a token can be sold, it will be — not necessarily because the recipient wants to sell it, but because poverty creates pressure that rationality cannot always resist. Soulbound tokens make that extraction architecturally impossible.

At the same time, locking everything in non-transferable tokens would be patronising and economically crippling. People need to save, to invest, to build toward long-term goals. The market layer handles this. Also the mSAWA layers allows for incentives structures and peg security mechanisms to kick off. The two layers coexist without compromising each other.

What a SAWA basket actually looks like

Because SAWA's unit of account is defined by what it buys, we need to be specific about what it buys. The SAWA basket covers eight domains of human need — domains that have emerged consistently across the major frameworks for understanding human welfare, from Amartya Sen's capability approach to the UN Sustainable Development Goals:

Nutrition and water: 63,000 calories per month through a least-cost nutritious diet meeting WHO and FAO standards. Staple grains, pulses, cooking oil, seasonal vegetables, safe water. Caloric minimums plus micronutrient adequacy — not just survival, but health.

Energy: 150 kilowatt-hours of electricity equivalent per month. Enough for lighting, phone charging, basic appliances, and food preparation. Grid-equivalent alternatives where the grid doesn't reach.

Communication: 10 GB of mobile data and 200 voice minutes. This is not a luxury in 2025. Communication is how people access healthcare information, send money home, find work, stay connected to family, and participate in civic life. It belongs in a dignity baseline.

Health and hygiene: Monthly supply of soap, toothpaste, menstrual products, and the equivalent of one primary healthcare consultation. Basic, not comprehensive — but the difference between this and nothing is the difference between dignity and preventable suffering.

Mobility: Minimum transport access to enable participation in work, education, and markets. Calibrated locally: a transit pass in São Paulo, a bicycle maintenance allowance in Kampala, something else elsewhere.

Education and culture: Learning materials and cultural participation. Because a human life involves more than biological maintenance.

Cognitive tools: This category reflects an emerging reality: access to AI-assisted information, decision support, and productivity tools is increasingly necessary to participate in economic and civic life on equal terms.

Civic participation: Support for exercising democratic rights and maintaining civic credentials. This one is deliberately unconventional — it reflects SAWA's commitment to the idea that participation in collective governance is a human need, not a luxury.

Each basket is globally structured but locally specified. The eight categories are universal; the specific goods, quantities, and optional additions are determined by the Citizens' Assembly for each zone. A basket in Dakar will specify local rice varieties, seasonal vegetables, and Senegalese mobile credit providers. A basket in Lyon will specify different staples, electricity contracts, and transit options. Both provide the same real capability guarantee, expressed in locally meaningful terms.

Amadou and Fatou: A Story About Remittances

Abstract monetary architecture becomes real when you put people in it. So consider Amadou and Fatou.

Amadou is a 34-year-old Senegalese construction worker living in Lyon. He earns €1,800 a month. Every month, he sends €250 home to his wife Fatou and their three children in Dakar. This is not unusual — Sub-Saharan African immigrants to France collectively send billions of euros home every year. This money is often the primary income source for entire extended families.

The process of sending that money is expensive, slow, and precarious. Western Union charges 7–10% on the transaction. The money arrives in CFA francs. Fatou converts it to cash and goes to the market. If the CFA is having a bad month — if French monetary policy has knocked the euro sideways, or if Senegalese food prices have spiked due to a drought in Mali — the food she can buy for her family is less than last month, even though Amadou sent the same amount. The value Amadou worked for has been eroded by the gap between what money claims to be and what it actually buys.

Now run the same scenario with SAWA.

Amadou purchases mSAWA tokens in Lyon at the current euro rate. He sends them to Fatou's wallet. Transaction fee: 1.5%. The tokens land in Dakar, where they're redeemable for exactly what the Dakar Citizens' Assembly has specified as the local basket equivalent. Fatou walks to a partner shop and redeems tokens for rice, cooking oil, a bag of vegetables, and mobile credit — quantities defined by the basket specification, immune to CFA franc fluctuations. The quantity of food Fatou receives is the same whether the CFA strengthened or weakened this month, because the tokens are anchored to the food itself, not to an intermediate currency.

Amadou saved €15–20 in fees. Fatou received a stable real-value transfer. The children ate the same amount as last month. For one family, this is meaningful. Scaled to the millions of remittance corridors that constitute a major component of economic survival across the Global South, it's transformative.

Remittance corridors are SAWA's ideal launch environment. There's inherent cross-border value before any local network is dense enough to sustain the system on its own.

The problem of making money work for Everyone

Here is the honest difficulty with SAWA, or with any proposal to redesign money at this scale. It requires trust before it can demonstrate trustworthiness. It requires adoption before it can demonstrate value. It requires merchants to accept it before consumers will use it, and consumers to use it before merchants will accept it. Every new monetary system faces this bootstrapping problem, and most fail there.

Local currencies have been trying to solve this for decades and have not managed to be used as actual money in any meaningful context. The problem is not the technology. It's that people use what other people use. Network effects in monetary systems are extraordinarily powerful precisely because money is only useful when other people accept it. This is what the rise of stablecoins has shown in recent years. Web3 technologies make such decentralized currencies possible, network effects make them usable.

SAWA's answer to this is the sequencing of its adoption strategy. It doesn't start with Universal Basic Provisioning. It starts with mSAWA — the market layer — in contexts where value is obvious immediately to participants: remittance corridors, where cross-border transfer savings are visible from day one; salary denomination for employees of participating organisations, who have an immediate reason to hold and spend the currency; and merchant networks in zones where the inflation problem is acute enough that stable real-value money is a compelling alternative to local currency.

Once mSAWA has built a functioning commercial ecosystem — once people are earning, saving, and spending it, once merchants have integrated it, once its stability has been demonstrated through real market conditions — then Universal Basic Provisioning becomes operationally feasible. Launching UBI before the network is ready would create redemption demand that outstrips provisioning capacity and collapse trust before it can be built. The sequence matters.

This is not a rushed proposal. The design envisions six phases over perhaps a decade: foundation infrastructure, small pilots in two or three cities, controlled regional expansion, salary denomination normalisation, then UBP introduction, and finally national and inter-national integration. The ambition is large; the pace is careful.

What gets replaced, and what doesn't

SAWA is designed to be complementary to existing monetary systems, not to replace them. Central banks retain their currencies. Governments retain their fiscal instruments. Tax is paid in the local currency. SAWA operates as an additional layer: a stable real-value instrument that citizens and organisations can choose to use alongside whatever they currently use.

This positioning makes SAWA politically feasible in ways that more radical monetary proposals are not. A proposal to abolish the euro or replace the dollar would mobilise every institutional power in the world against it before it got started. SAWA doesn't ask permission from the ECB or the Fed. It asks permission from supermarkets to accept a new payment instrument, and from people to try a currency that actually holds its value.

There is something historically significant in this approach. The major monetary innovations of the last century — Eurodollars, mobile money in Kenya, PayPal, stablecoins — did not begin by replacing existing systems. They began by doing something specific that existing systems did badly. M-Pesa solved rural mobile payments in Kenya not by becoming a central bank but by being more useful than what existed for a specific population with a specific need.

SAWA's specific need is essential-goods inflation protection. If it does that well adoption will follow. Not because it's ideologically compelling, but because it's practically useful.

The deeper question: what is money for?

All of this — the architecture, the governance, the baskets, the redemption mechanics — is in service of a philosophical claim that deserves to be stated plainly.

Money is a social technology. Like all technologies, it was designed by human beings for human purposes, and it can be redesigned. The question is not whether money can be different. It clearly can be — it has been many different things over the past five thousand years, from commodity barter to cowrie shells to gold coins to central bank notes to digital ledger entries. The question is what we want money to do.

The current answer, operationalised in every major monetary system on earth, is something like: money should enable efficient exchange, store value well enough to be useful, and maintain enough stability to underpin contracts and planning. These are reasonable goals. But they are institution-centred goals. They describe what money should do for banks, corporations, and governments. They say very little about what money should do for the person earning minimum wage, or the retired teacher, or the subsistence farmer, or the urban refugee.

SAWA's answer is different: money should preserve human capability. Not just financial value in the abstract — human capability, concretely understood as the ability to eat, to stay warm, to communicate, to participate in the life of one's community. A monetary system built around this definition would anchor its unit to the things that make human life possible, govern its basket democratically, make its guarantee physically enforceable, and ensure that its floor cannot be sold out from under the people who most need it.

The question of how many months of decent life a person can hold in their hands — that is the measure of things that counts.

Whether SAWA achieves this is an empirical question that will require pilots, stress tests, and years of evidence. The full research document includes formal models of redemption dynamics, liquidity sufficiency, peg stability under crisis conditions, adoption diffusion, and risk mitigation — the kind of careful, falsifiable analysis that monetary proposals require before they can be taken seriously.

But before the equations, there is the animating conviction. Money has been built to serve institutions. SAWA proposes to build money that serves people. Not as a welfare transfer — as a redesign of the fundamental unit of measurement. A currency whose anchor is human need rather than institutional promise.

Why this might actually happen

Proposals to redesign money are not rare. Most of them go nowhere. Why might SAWA be different?

Three structural conditions have converged in the past decade that make a project like SAWA not just imaginable but arguably necessary.

The first is the inflation shock of 2021–2024, which reminded populations in wealthy countries — not just in the Global South — what it feels like when money doesn't protect you from rising essential costs. The political consequences are still unfolding. But the intellectual consequence is a reopened question: what is the point of price stability if it doesn't stabilise the prices that matter to people's lives?

The second is the maturation of digital payment infrastructures. M-Pesa in Kenya, UPI in India, Pix in Brazil have demonstrated that financial inclusion can be achieved at scale without relying on traditional banking infrastructure. The populations most exposed to essential-goods inflation are precisely the populations that have adopted mobile money most rapidly. SAWA can deploy on the infrastructure they already use. Further, Web3 technologies have created a credibly neutral, decentralized, byzantine tolerant, privacy preserving set of tools that allow for cheap, secure, decentralized money designs and distribution.

The third is the demonstrated feasibility of Citizens' Assemblies for complex governance decisions. A decade ago, the idea that randomly selected citizens could deliberate meaningfully on constitutional amendments or climate policy would have been considered naive. The evidence from the EU, France, Germany, and elsewhere suggests otherwise. The institutional model that SAWA requires for basket governance has been stress-tested.

None of this guarantees success. Monetary innovation is hard, slow, politically fraught, and frequently captured by the interests it was designed to challenge. The history of monetary reform is littered with ingenious proposals that solved the right problem and went nowhere.

But, to my knowledge, the problem that SAWA is solving — that the unit of money is not anchored to anything a human being can actually depend on — has not been solved by any existing system. And the consequences of leaving it unsolved are not abstract. They show up every day in Lagos and Lyon, in Dakar and Dhaka, in every household where the monthly arithmetic of food and energy and connectivity no longer adds up the way it used to.

Money should be worth something real. That seems like a modest requirement. It has never been fully met.

A Final Word

Imagine Mariama, a 42-year-old school cook in Dakar, receiving her first monthly pSAWA allocation in 2032.

She doesn't receive cash. She receives a notification on her phone: her token wallet has been topped up with the Dakar basket equivalent for the month — rice, oil, vegetables, mobile credit, hygiene supplies. She walks to the nearest partner shop, taps her phone, and receives the goods. The quantities are what the Dakar Citizens' Assembly decided, informed by nutritional science and local custom, not by what a central bank finds convenient.

If the price of rice doubles next month, Maria's basket doesn't shrink. The basket is the unit. The price adjustment happens in fiat, not in what she receives. Her floor is stable.

This is not a large scene. It contains no drama, no revolution, no declaration. It is simply a woman receiving what the currency promised. What is remarkable about it is how rarely, in the long history of monetary systems, that has happened.

Populism wins not by delivering better policy but by providing what liberal democracies have stopped offering: identity, belonging, meaning, recognition. This isn't irrational—it reflects our evolved nature. Humans are wired for face-to-face cooperation; physical co-presence triggers oxytocin release and neural synchronization essential for trust and collective action. Digital spaces cannot replicate this, and now face an authentication crisis as AI floods online discourse with synthetic participation. The democratic response is a politics of presence: physical citizens' assemblies where real humans gather in real time to deliberate on real questions. These spaces generate both considered judgment and genuine community—the same neurobiological bonding that populist rallies exploit, but channeled toward cooperative rather than antagonistic ends. Assemblies are not nostalgic attempts to restore lost community but modern institutions for autonomous individuals who nonetheless need belonging. The body politic, reassembled.

The paradox of populist persistence

Something puzzling is happening in democracies around the world. Populist leaders who fail to deliver on their promises—who preside over economic decline, institutional erosion, and broken commitments—nonetheless retain substantial political support. This pattern defies the conventional wisdom that voters evaluate incumbents based on performance. If citizens are rational actors who reward delivery and punish failure, populism should be self-limiting: once in power, populists would be held accountable for their results. Yet the evidence tells a different story.

Consider the empirical record. Funke, Schularick, and Trebesch's landmark 2023 study in the American Economic Review analyzed 51 populist leaders across 60 countries from 1900 to 2020. Their findings are striking: after fifteen years, GDP per capita is approximately ten percent lower under populist governments compared to non-populist counterfactuals. Populist rule is consistently associated with economic disintegration, decreasing macroeconomic stability, and institutional erosion. Yet this economic damage does not automatically translate into electoral defeat.

The case studies are equally instructive. Turkey's Erdoğan won re-election in 2023 amid inflation exceeding eighty percent. Mexico's López Obrador maintained extraordinary popularity despite a weak economic record, and his hand-picked successor won by a record margin. Hungary's Viktor Orbán has retained power for over a decade despite persistent corruption allegations. Silvio Berlusconi governed Italy on and off for two decades despite numerous scandals. In Venezuela, the Chávez-Maduro regime transformed a wealthy oil exporter into an economic disaster while maintaining significant support, at least initially.

Research on voter loyalty confirms this persistence. Rooduijn, van der Brug, and de Lange, writing in Government and Opposition in 2016, found that populist voters are as loyal as mainstream party voters. Contemporary populist parties have proven remarkably adept at establishing themselves as permanent outsiders despite being fully integrated into political systems.

This persistence demands explanation. If populism regularly fails to deliver material benefits—and often actively harms its supporters economically—why does support endure? The answer lies in recognizing that populism succeeds through a fundamentally different mechanism than conventional politics assumes.

Inputs over outputs: reframing the appeal of populism

The conventional model of electoral accountability treats voting as a mechanism for rewarding or punishing performance. Citizens evaluate what governments deliver—economic growth, public services, culture, security—and vote accordingly. This model is elegant but incomplete. It cannot explain why voters would continue supporting leaders whose policies demonstrably harm them.

A growing body of research suggests we need to distinguish between inputs and outputs in political support. Outputs are the tangible deliverables of governance: policy implementation, economic performance, public goods provision. Inputs are something different: the psychological and social goods that political participation itself provides, including identity, belonging, meaning, recognition, and community.

Daniel Brieba, writing in the Political Quarterly in 2025, argues that voters make choices based on social identities and support candidates with whom they identify, often irrespective of policy positions or outcomes. This is not irrationality but a different kind of rationality—one oriented toward expressive rather than instrumental goals.

The concept of "nostalgic deprivation" is particularly illuminating. Research by Gest, Reny, and Ferwerda, surveying 20,000 individuals across 19 European countries, found that voters support populists because they believe people like them had it better in the past. This is not primarily about economic decline but about status, recognition, and social position.

Rico, Guinjoan, and Anduiza, in their 2020 review in Current Opinion in Psychology, demonstrate how populist movements tap into deep psychological needs for belonging and status. Social identity theory suggests that identity threat caused by inter-group comparison is a crucial variable explaining attraction to populist ideas. Populism offers not just policies but a place in a meaningful collective—an "us" defined against a "them" that provides clarity, purpose, and belonging.

Salmela and von Scheve's work on the emotional roots of populism reinforces this picture. Shame, fear, and resentment—emotions tied to perceived loss of status and recognition—drive populist support more powerfully than calculations of material interest. Populist movements transform these negative emotions into positive collective identity.

If this analysis is correct, the implications for democratic response are profound. Competing with populism on outputs alone—better policy, stronger economic performance—may be necessary but is unlikely to be sufficient. Liberal democracy must also provide inputs: identity, belonging, meaning, and recognition.

What liberal democracy lost—and gained: the transformation of democratic life

It would be a mistake to conclude that liberal democracy is inherently incapable of generating belonging and meaning. This is historically inaccurate. For much of the nineteenth and twentieth centuries, democratic participation was embedded in rich organizational ecosystems that provided exactly what populism now offers—and more.

Consider the mass parties that dominated democratic politics from the late nineteenth century through the mid-twentieth century. The German Social Democratic Party (SPD), British Labour, Christian Democratic parties across Europe—these were not merely electoral machines but total social ecosystems. They operated youth wings and women's organizations, sports clubs and cultural associations. They published newspapers and organized reading circles, educational associations and mutual aid societies. They provided funeral societies and cooperative housing, summer camps and festivals and parades. To be a Social Democrat or a Christian Democrat was not merely to vote a certain way but to inhabit a way of life.

Democratic movements have similarly demonstrated the capacity to generate profound meaning and belonging. The American civil rights movement, Polish Solidarity, the anti-apartheid struggle in South Africa, the Indian independence movement—these created communities of extraordinary depth and purpose. Participation was transformative.

What happened to these organizational ecosystems? The story is not simply one of decline. The same decades that saw the hollowing out of mass parties also witnessed genuine liberation: expanded rights for women, minorities, and LGBTQ+ people; rising education levels enabling independent judgment; greater individual autonomy and freedom from conformist social pressure; the pluralization of lifestyles and identities. The old forms of belonging came bundled with constraints that few would wish to restore. You were a Social Democrat because your neighborhood, workplace, and family were—not necessarily because you chose it freely. These communities could be patriarchal, conformist, and coercive.

Yet something was lost alongside these gains. Peter Mair's Ruling the Void documents how parties became professionalized and cartel-like, floating above civil society rather than embedded within it. The infrastructure for collective meaning-making atrophied without replacement. Modern individuals are freer but also more atomized, more educated but often lonelier, liberated from coercive community but hungry for authentic connection. Chantal Mouffe has argued that liberal democracy compounded this by trying to eliminate passion and antagonism from politics in favor of technocratic consensus—draining politics of emotional meaning and leaving a vacuum that populism eagerly filled.

The challenge, then, is not to recreate the old mass parties—with their conformism and inherited identities—but to build new forms of democratic belonging compatible with modern pluralism and individual autonomy. Citizens' assemblies may be particularly suited to this task: participation is chosen rather than inherited, random selection brings together strangers rather than tribal in-groups, engagement is time-limited rather than total, and deliberation respects individual judgment while creating genuine community. The task is not nostalgic restoration but forward-looking reinvention.

The neurobiological foundations: why physical presence matters

Before turning to institutional design, we must understand why physical assembly specifically—rather than any form of collective participation—is essential. The answer lies in our evolutionary heritage and the neurobiology of human sociality. We are not merely social by preference but social by biological design, and our brains require physical co-presence to function as social organs.

Robin Dunbar's "social brain hypothesis," developed through decades of comparative research, provides the evolutionary foundation. Primates evolved unusually large brains—specifically enlarged neocortices—to manage the complexity of social life. There is a quantitative relationship between neocortex size and social group size across primate species: the larger the brain, the larger and more complex the social groups the species can sustain. Extrapolating from this relationship to human brain size yields the now-famous "Dunbar's number" of approximately 150—the cognitive limit on the number of stable social relationships a human can maintain.

Crucially, this research suggests that human intelligence is fundamentally social intelligence. We did not evolve large brains primarily to solve ecological problems like finding food or making tools. We evolved them to navigate the challenges of living in complex cooperative groups: forming alliances, tracking relationships, managing reputations, coordinating collective action. Our cognitive architecture is optimized for face-to-face social coordination.

Rutger Bregman's Humankind: A Hopeful History synthesizes evidence from archaeology, anthropology, and psychology to argue that this social nature is fundamentally cooperative rather than competitive. Drawing on the philosophical debate between Rousseau (humans are naturally good) and Hobbes (humans are naturally violent), Bregman marshals evidence that the Rousseauian view is closer to the truth. Humans are not self-interested individuals held in check by civilization but cooperative beings whose natural sociality is often suppressed rather than enabled by modern social arrangements.

This has profound implications. If cynical views of human nature become self-fulfilling prophecies—if institutions designed on the assumption that people are lazy or selfish bring out those behaviors—then institutions designed on assumptions of human cooperativeness might bring out our better natures. Deliberative processes , designed to trust participants with genuine responsibility and meaningful deliberation, may succeed precisely because they treat citizens as the cooperative beings we evolved to be.

The neurobiological mechanisms underlying human sociality have been extensively documented. Neurobiologist Nicole Strüber's recent work Unser soziales Gehirn (Our Social Brain) synthesizes research on the biochemistry of togetherness. Physical co-presence triggers the release of oxytocin and other neurotransmitters that promote relaxation, empathy, trust, and openness to change. Strüber emphasizes a key concept: synchronization. When humans interact face-to-face, their brains literally synchronize—neural oscillations align, autonomic responses coordinate, even heartbeats can entrain. This synchronization does not happen—or happens far less effectively—through digital mediation.

Research on oxytocin and interpersonal coordination confirms this picture. Studies using EEG hyperscanning—simultaneously recording brain activity from interacting individuals—show that synchronous social interactions evoke heightened oxytocin release in both partners, which in turn enhances neural synchrony in regions involved in social cognition and emotion processing. Oxytocin administration increases interpersonal coordination, and synchronization promotes cooperation, trust, and affiliation in a reinforcing cycle.

The implications are striking. The emotional bonds and collective identity that populism generates through rallies and spectacles are not arbitrary—they tap into evolved mechanisms for group bonding. But these same mechanisms can be engaged by democratic participation, if that participation involves physical co-presence. WhatsApp messages do not release oxytocin the way face-to-face conversation does. Video conferences do not produce neural synchronization the way sitting together in a room does.

Strüber's diagnosis is pointed: "We spend ever less time in genuine togetherness—WhatsApp messages instead of spontaneous visits, video conferences instead of in-person meetings, efficiency-driven family life, medical appointments run by stopwatch. Our brain needs this exchange, however. We synchronize, and neurotransmitters like oxytocin are released."

Dunbar's research confirms that digital communication has not expanded our social capacities. Despite the theoretical possibility of maintaining unlimited online connections, actual social network sizes remain constrained by the same cognitive limits that applied before the internet. The layered structure of human social networks—with an innermost circle of about 5 close relationships, then 15, then 50, then 150—appears in both traditional societies and online networks. The glass ceiling on genuine social connection remains where it always was, because it reflects the architecture of the social brain, not the limitations of communication technology.

The digital mirage: Why Digital deliberation cannot save democracy

An obvious response to the challenge of scale is digital technology. If the problem is connecting citizens to democratic participation, the internet would seem to offer unlimited potential. Online deliberation platforms, social media engagement, AI-facilitated discussion—surely these tools can rebuild democratic community?

The numbers are seductive. As of 2024, 5.41 billion people use social media worldwide—63.9 percent of the global population. Average daily usage exceeds two hours. If even a fraction of this engagement could be channeled toward democratic deliberation, the scale would dwarf anything achievable through physical assemblies.

Yet digital deliberation faces two insurmountable problems. The first, just discussed, is neurobiological: online interaction cannot generate the synchronization, oxytocin release, and neural coordination that physical presence and interaction produces. Digital democracy might transmit preferences but cannot build community in the neurobiologically meaningful sense.

The situation is actually worse than mere absence of bonding mechanisms. Social media platforms are engineered to activate an entirely different neurochemical profile—one optimized for engagement rather than connection. Research demonstrates that social media exploits the brain's dopamine reward system in ways analogous to slot machines and addictive substances: unpredictable rewards (likes, comments, notifications) trigger dopamine surges that create compulsive use patterns. As psychiatrist Anna Lembke explains, users enter a "dopamine deficit state" between sessions, experiencing reduced pleasure from ordinary activities and driving them back to the platform. Simultaneously, the constant stress of social comparison and fear of missing out elevates cortisol levels, which paradoxically amplifies dopamine sensitivity—creating addiction-stress cycles that platforms profit from but users suffer through. A 2024 neurobiological meta-analysis found that excessive social media use is associated with decreased serotonergic activity and increased cortisol release, "ultimately weakening emotional resilience and cognitive control."

More troubling still, these engagement-maximizing algorithms actively promote polarization. A November 2025 study published in Science found that exposure to antidemocratic and hostile partisan content shifted users' feelings toward the opposing party by two points on a 100-point scale—an effect normally observed over three years—demonstrating that algorithms causally alter affective polarization. Research on Twitter showed that moral-emotional language and outgroup derogation correlate with greater engagement.

Where physical co-presence releases oxytocin and promotes trust, social media triggers dopamine spikes and cortisol stress. Where face-to-face deliberation builds understanding across difference, algorithmic feeds amplify hostility toward outgroups. The neurobiological infrastructure of digital platforms is not merely inadequate for democratic community—it is actively corrosive of it.

The second problem is the authentication crisis which is not a technical limitation to be overcome but a structural feature that will only worsen as artificial intelligence advances.

The problem operates at multiple levels. First, there is simple attestation failure: in any online forum, how do we know that a contribution comes from a real human being rather than a bot or AI system? Second, even if we could verify human initiation, there is delegation ambiguity: did the human meaningfully engage with the question, or did they simply deploy an AI agent to participate on their behalf? Third, and perhaps most troubling, is what might be called synthetic drift. Large language models are increasingly trained on data that includes AI-generated content. The outputs of these models then enter the information ecosystem and become training data for future models. We are approaching a point where AI systems optimize for patterns that are themselves AI-generated, creating feedback loops with uncertain democratic legitimacy. Fourth, there is adversarial pollution. Bad actors—whether state adversaries, commercial interests, or ideological movements—can flood online deliberation systems with synthetic participation at minimal cost. The economics are asymmetric: authentic participation requires genuine human time and attention, while synthetic participation can be generated at scale.

Finally, all of this leads to legitimacy collapse. If citizens suspect that input to democratic processes is substantially synthetic, why should they trust the outputs?

This analysis suggests that the digital solution is not merely inadequate but actively counterproductive as a primary mode of democratic participation: It doesn't solve the belonging problematic and may lead to even less trusted outputs . What is needed is something that can create belonging at scale while producing sound policy results.

Physical assemblies as democratic infrastructure of the future

A body in a room, contributing in real time, in synchronicity with other bodies and minds—this is the one form of participation that resists synthetic replication and creates belonging. Physical presence provides what online engagement cannot: unforgeable authentication, witnessed identity, real-time accountability, embodied commitment, and readable social cues. And, crucially, it provides what our evolved social brains require: synchronization, oxytocin release, and genuine community formation.

Physical assemblies are of course not a novel proposal. They are the oldest form of collective decision-making, present in virtually every human society across history. The councils of countless traditional societies, the village assembly in India, the Athenian ekklesia, the Roman comitia, the Icelandic Althing, the New England town meeting, the Swiss Landsgemeinde—all share a common core: citizens gathering in physical space to deliberate and decide together. Before there was writing, before there were constitutions, before there were political parties, there were assemblies. The practice is as old as politics itself, because it is rooted in what we are as a species: social animals who evolved to coordinate face-to-face.

Even today, physical assemblies remain central to political life in ways we often overlook. Parliaments and councils meet in chambers. Juries deliberate in rooms. Shareholders convene in halls. Trade unions hold congresses. Social movements gather in squares. Religious communities assemble in churches, mosques, and temples. The moments that matter most in collective life—the decisions, the rituals, the transformations—still happen when people gather in shared physical space.

In recent decades a newcomer in the world of assemblies as gained traction under various names—planning cells, citizens' dialogues, citizens' forums, citizens' panels, citizens' juries, citizens' assemblies, mini-publics, deliberative polls—a family of practices has emerged that share common design principles. As a note, the term "citizens' assemblies" has become particularly prominent—perhaps overly so. It is important to recognize that this is not the only form deliberation can take, and the label matters less than the underlying mechanisms. What makes these practices distinctive is a specific combination of design choices that together create something genuinely new:

Sortition. Participants are selected by lottery from the general population, often stratified to ensure demographic representativeness. This is the crucial innovation that distinguishes these practices from traditional town halls, public hearings, or activist gatherings. Random selection means that participants are not self-selected enthusiasts, partisan mobilizers, or professional stakeholders. They are ordinary citizens who would never otherwise have met, brought together by chance. This has profound implications for belonging: unlike churches or parties, where membership follows pre-existing identity (you join people like yourself), sortition-based assemblies forge connection across difference. Strangers become collaborators. The belonging that emerges is not tribal but civic.

Structured deliberation. Participants do not simply show up and argue. They receive balanced information from diverse experts, work through issues in facilitated small groups, hear from affected stakeholders, and build toward considered judgment over multiple sessions. The process is designed to enable genuine learning and mutual understanding, not merely the aggregation of pre-formed preferences.

Physical co-presence. Participants meet in person, in real time, in shared space. As discussed earlier, this triggers the neurobiological mechanisms—synchronization, oxytocin release, emotional attunement—that digital interaction cannot replicate. The embodied experience of working through difficult questions together creates bonds that persist beyond the assembly itself. Of course in many cases, physical session of work alternate with online synchronous session to leverage the advantage of the online setting.

Synchronous engagement. Everyone participates in the same process at the same time. This shared temporality—the knowledge that others are grappling with the same questions in the same moments—creates collective rhythm and mutual accountability.

This combination of mechanisms produces something that traditional political institutions struggle to generate: belonging that is compatible with modern pluralism. Like churches or mass parties, these assemblies create genuine community through shared practice. But unlike those older forms, they do not require prior ideological commitment or inherited identity. They bring together people who disagree, who come from different backgrounds, who would never encounter each other in their ordinary lives—and ask them to work together on questions that matter. The belonging that emerges is chosen rather than given, forged rather than inherited, and oriented toward shared problem-solving rather than tribal solidarity against outsiders.

The track record demonstrates that such processes produce high-quality results on genuinely difficult issues. Ireland's Citizens' Assembly addressed abortion and marriage equality—questions that had paralyzed conventional politics for decades—and produced recommendations that led to successful constitutional referenda. France's Citizens' Convention on Climate generated 149 proposals, approximately 70 percent of which were adopted by the government. The European Citizens' Panels have continuously fed into European policymaking since 2022, with participants from across member states deliberating on topics from food waste to virtual worlds. Thousands of local participatory budgets, planning panels, and deliberative forums operate worldwide, giving citizens direct voice in decisions that affect their communities. In terms of belonging participants clearly express how the experience kicks off a very deep change: a large portion of participants in such process declare that their feeling of trust and belonging towards democracy and institutions increases through the participation. As many participants of European Citizens' Panels put it: "now I feel European".

So, there is strong evidence that such assemblies could function as the foundation of a renewed democratic infrastructure. To do so, they need to operate at a different scale. They must become something that every citizen experiences, not something that happens to a lucky few.

Scaling Deliberation: Four dimensions of the democratic infrastructure

How can deliberative assemblies operate at the scale required to serve as genuine infrastructure for democratic belonging? The Scaledem project, a Horizon Europe initiative examining how to move democratic innovations beyond isolated experiments, identifies four critical dimensions of scaling that any comprehensive strategy must address: scaling high (embedding in governance structures), scaling out (reaching more people), scaling in (improving procedural quality), and scaling deep (meaningfully engaging and creating belonging).

For citizens' assemblies to address the populist challenge, they must succeed on all four dimensions simultaneously. Scaling high without scaling deep produces technocratic exercises that generate recommendations but not community. Scaling out without scaling in sacrifices deliberative quality for mere participation numbers. The distinctive contribution of this analysis is to argue that scaling deep—creating genuine belonging and meaning-making—is not a luxury to be added once the logistics are solved but rather the essential foundation without which the entire enterprise fails to address the challenge of populism.

Scaling Deep: The foundation of democratic belonging

Scaling deep means creating the conditions under which the neurobiological mechanisms described earlier can operate fully—where participants not only deliberate on policy but experience the synchronization, oxytocin release, and neural coordination that transform strangers into a community. This requires more than structured discussion; it requires time and space for genuine human encounter.

Concretely, scaling deep might involve dedicating one full day of a multi-day assembly process entirely to mutual exchange and sense-making activities. Participants could share something meaningful to them: personal stories, music that moves them, books that shaped their thinking. They might cook together, sing together, dance together, or engage in physical activities as a group. The point is not to be productive in any policy sense but simply to be with one another—to allow the slow work of human bonding that our social brains evolved to accomplish but that modern life so rarely permits.

This is not frivolous. The research on interaction ritual chains suggests that precisely such shared experiences—marked by physical co-presence, common focus, emotional attunement, and rhythmic coordination—generate the "emotional energy" that participants carry forward into substantive deliberation. The neurobiological research on synchronization confirms that these effects are not merely subjective: brains literally align, oxytocin flows, and the physiological preconditions for trust and openness emerge. Policy work conducted after such bonding draws on reserves of goodwill and mutual understanding that cold deliberation alone cannot generate.

Scaling Out: Decentralization, sequencing Rotation

Scaling out involves three complementary mechanisms that together can bring deliberative participation within reach of every citizen while addressing questions at every scale of governance.

Decentralization with integration

The crucial insight is that decentralization does not mean assigning local topics to local assemblies and global topics to global ones. Rather, it means enabling any question—including planetary ones—to be deliberated in thousands of local settings simultaneously, with results flowing upward or transversely through structured integration processes.

The World Wide Views and Global Citizens' Dialogue model pioneered this architecture: in 2015 ten thousand citizens across seventy-six countries deliberated on climate policy on the same day, using common information materials and guiding questions, producing recommendations that were then compiled into a global synthesis presented to the Paris climate negotiations. We the Internet convened 5,500 participants across more than seventy countries in October 2020 to deliberate on digital governance.

This architecture preserves the neurobiological benefits of many small, face-to-face local deliberation where synchronization, oxytocin release, and genuine human bonding can occur; meaningful connection between local participants and global outcomes; and the production of recommendations that carry the legitimacy of broad-based citizen input.

At least 3 architectural patterns can integrate local deliberations into higher-level synthesis:

The ambassador model: Local assemblies deliberate on a shared question and then delegates—"ambassadors"— are randomly selected to participate in a higher level assembly that works across the local productions. These ambassadors carry the mandate of their local assembly but also encounter the perspectives of ambassadors from other localities. The centralized assembly does not start from scratch but builds upon the foundation laid by parallel local deliberations. The ambassador is not a representative in the electoral sense—not someone empowered to substitute their judgment for the group's—but a carrier of the local assembly's collective reasoning into a higher-level synthesis. This model adds a second layer of deliberative experience: the ambassador who deliberated locally with neighbors then deliberates transnationally with fellow ambassadors, experiencing both scales of democratic community.

The synthesis model: Local assemblies produce recommendations that are aggregated, clustered, and synthesized by a higher level assembly without requiring physical delegates from each locality. The centralized assembly— itself composed of randomly selected citizens who did not participate in the local phase—works with the distilled outputs of the local deliberations as their evidentiary base, much as traditional assemblies work with expert testimony. This model allows the centralized assembly to benefit from the breadth of local input while maintaining its own deliberative integrity.

The transversal model: Assemblies working on similar topics in different contexts can exchange insights horizontally, learning from one another without requiring vertical integration through a higher-level synthesis. A citizens' assembly on forestry policy in Sweden—grappling with boreal forest management, timber industry interests, and Indigenous Sámi land rights—could learn enormously from an assembly on forestry in Ghana addressing tropical deforestation, cocoa farming pressures, and traditional land tenure systems. The ecological, economic, and cultural contexts differ profoundly, yet each assembly confronts the fundamental challenge of balancing human use with ecological sustainability, and each has developed solutions the other might adapt.

Transversal exchange can take several forms: assemblies might share their recommendations and background materials, allowing each to consider how a different context approached similar challenges; delegate exchanges might bring participants from one assembly to observe or briefly join another's deliberations; joint digital sessions might connect assemblies in real-time for structured dialogue; or a shared platform might compile approaches from assemblies across multiple contexts, creating a living repository of citizen-generated policy wisdom.

This horizontal architecture generates a distinctive form of belonging: solidarity across difference. The Swedish forest assembly participant who encounters the reasoning of Ghanaian citizens—and recognizes both the vast differences in context and the shared commitment to thoughtful collective judgment—experiences democratic community that transcends not only national borders but ecological zones, economic systems, and cultural traditions. The recognition that citizens everywhere are engaged in the same practice of deliberative self-governance, adapting it to radically different circumstances, builds a sense of global democratic citizenship that vertical integration alone cannot produce.

Transnational belonging through shared deliberation. This architecture addresses the parochialism concern in a way that purely local deliberation cannot. The participant in a neighborhood assembly in Lyon deliberating on global climate policy, aware that citizens in Lagos and Lima are engaged in the same question at the same moment using the same information base, experiences herself as part of a planetary democratic community even without leaving her arrondissement. When her assembly's ambassador travels to the global synthesis assembly—or when the Lyon recommendations appear alongside those from Lagos and Lima in the final synthesis document—the connection becomes concrete.

The European Citizens' Panels demonstrate that transnational deliberation generates transnational identity. As the Observatory Report on the 2022-2023 panels documented, participants spoke of developing "their greater understanding of their own 'European' identity." This effect can be multiplied when thousands of local assemblies across a continent or the globe deliberate on the same question simultaneously: the shared temporal experience of collective deliberation—knowing that at this very moment, fellow citizens across borders are wrestling with the same challenges—creates bonds that transcend the physical limits of any single assembly room.

Sequential phasing for depth and continuity

Single topics can unfold across multiple assembly cohorts, with each phase building on the previous. Germany's Bürgerrat Bildung und Lernen provides a template: over seven hundred citizens participated across multiple phases—scoping, diagnosis, vision, objectives, and recommendations—with partial rotation ensuring both continuity and fresh perspectives.

Participant rotation for universal reach

With rotation, every citizen could expect to serve in a deliberative assembly at least twice over their active citizen life—from childhood through old age, roughly ages six to eighty-five. For a country like Germany with approximately seventy million adults, achieving two participations per citizen over a fifty-year planning horizon would require roughly 2.8 million participant-slots annually. This is achievable through a distributed architecture operating at multiple scales simultaneously:

• Global level: Approximately 100 synthesis assemblies per year, each integrating input from thousands of local assemblies across countries. Topics would include climate, migration, technology governance, pandemic preparedness, nuclear security, and other challenges requiring planetary coordination. A Global Citizens' Assembly on climate, for instance, might draw on parallel local deliberations in every UN member state, with ambassadors or synthesized recommendations flowing to a centralized deliberation that complements intergovernmental processes.

• Regional level: Approximately 50 synthesis assemblies per year across major regional bodies—European Union, African Union, ASEAN, Mercosur, and others. The EU currently runs two to four European Citizens' Panels annually; scaling to approximately ten per year while other regional organizations establish similar processes would build regional democratic community. Each regional assembly would integrate local deliberations from across member states on questions of cross-border significance.

• National level: Approximately 50-100 synthesis assemblies per year per country, roughly aligned with major legislative initiatives. With the ambassador or synthesis model, each national assembly would integrate input from hundreds of local assemblies distributed across the territory—ensuring that citizens in rural areas, small towns, and underrepresented regions have equal voice with those in capital cities.

• Local entry points: The base layer consists of thousands of local assemblies—at neighborhood (15-30 participants), district (30-50), and city (50-100) levels—that serve as the primary sites of citizen deliberation. These are not limited to "local issues" but deliberate on questions at every scale, producing recommendations and selecting ambassadors that flow upward into synthesis processes. A neighborhood assembly in a German village deliberates on the same EU digital policy question as one in central Paris; both feed into national syntheses that feed into the European Citizens' Panel.

The optimal sizes at each level—15, 30, 50, 100, 150—correspond to the layered structure of human social groups identified by Dunbar's research on the social brain. These are not arbitrary administrative divisions but reflect the cognitive architecture that evolution has given us for managing social relationships. Assemblies sized to these natural groupings can leverage the full capacity of human social cognition.

The math

Top-down: If approximately 100 global, 50 regional, and 50-100 national synthesis assemblies operate annually, and each integrates input from 500-1,000 local assemblies (each with 30-50 participants), the system generates between 2.5 and 10 million participant-slots per year at the local level alone, plus additional slots for synthesis assembly participants and ambassadors.

Bottom-up: If the goal is two deliberative experiences per citizen over a seventy-eight-year active citizen life (ages 6-85), a country of seventy million people requires approximately 1.8 million participant-slots annually. Distributed across local assemblies feeding into higher-level syntheses, this translates to roughly 36,000-60,000 local assemblies per year (at 30-50 participants each), or about 100-165 per day—entirely feasible for a country with thousands of municipalities, schools, workplaces, and community organizations that could host deliberative processes.

The two calculations converge: a system designed to address the major policy questions facing humanity at global, regional, and national levels, using decentralized deliberation with upward synthesis, would generate approximately the right number of participation opportunities to ensure that every citizen deliberates at least twice over their lifetime.

Scaling In: Integrating policy work with meaning-making

The fourth dimension—scaling in—concerns procedural design quality. For our purposes, the crucial insight is that policy deliberation and meaning-making should not be separate activities but integrated dimensions of a single practice.

Traditional deliberative design often treats the "substantive" work of producing recommendations as primary and any community-building activities as optional add-ons. This gets the relationship backward. Without the trust, empathy, and mutual understanding that emerge from genuine human encounter, policy deliberation risk to degrade into positional bargaining among strangers. The recommendations produced may be technically sophisticated but will lack the legitimacy that comes from having been forged by a genuine community.

Design elements that integrate belonging into deliberation include:

• Opening rituals that mark the transition from ordinary life into deliberative space and acknowledge the significance of the collective task

• Storytelling rounds in which participants share personal experiences related to the policy topic before engaging in abstract discussion

• Creative expression opportunities—collective art-making, music, movement—that activate non-verbal channels of human connection

• Shared meals prepared and consumed together, tapping into the ancient human practice of commensality as community-building

• Closing ceremonies that acknowledge what has been accomplished together and create transitional objects (photographs, collective statements, small gifts) that participants carry back into ordinary life

• Continuity elements such as reunions, alumni networks, and follow-up communications that maintain relationships beyond the assembly period

Randall Collins's theory of interaction ritual chains suggests that successful rituals generate "emotional energy" that participants carry forward, motivating continued engagement and spreading positive affect through social networks. Durkheim's concept of collective effervescence captures the same phenomenon: the emotional electricity that emerges when humans gather with shared focus and common purpose. The neurobiological research confirms that these are not merely metaphors but descriptions of real physiological processes. The goal is not to manufacture artificial enthusiasm but to create conditions under which the natural human capacity for bonding can operate. When participants return from a well-designed assembly, they should carry not only policy recommendations but transformed relationships—with fellow participants, with the democratic process, and with the larger political community of which the assembly was a microcosm.

What role for artificial intelligence?

The earlier critique of digital deliberation emphasized AI's role in the authentication crisis—synthetic participation polluting online democratic spaces. But AI can also serve as technological infrastructure that makes scaled physical deliberation possible. Multilingual real-time translation enables assemblies to connect across language barriers, allowing the Swedish forestry assembly to engage directly with deliberators in Ghana. Synthesis tools can aggregate and cluster recommendations from thousands of local assemblies, identifying patterns of convergence and divergence that human analysts could not process at scale. Curated fact-checking and information provision can ensure that all assemblies work from reliable, balanced evidence bases. The distinction is crucial: AI as substitute for human deliberation undermines democratic legitimacy; AI as support for human deliberation enhances democratic capacity. The deliberation itself—the face-to-face encounter, the trust-building, the collective judgment—must remain irreducibly human. But the infrastructure connecting, translating, and synthesizing human deliberation across scales can and should draw on AI's capabilities.

Conclusion: Recovering democratic community

The argument can be summarized as a logical chain. Populism succeeds primarily through inputs—identity, belonging, meaning, recognition—rather than outputs. Liberal democracy historically provided these inputs through mass parties and social movements, but modern societies have undergone a double transformation: genuine gains in autonomy, education, and individual freedom, alongside increased atomization and the erosion of collective meaning-making. This leaves an emotional vacuum that populism fills—and this vacuum also contradicts our evolved nature: humans are fundamentally social beings whose brains developed for face-to-face cooperation, and physical co-presence triggers neurochemical processes essential for trust, bonding, and collective action. Digital deliberation cannot fill this gap—both because online interaction cannot generate the neurobiological effects of physical presence and because an insurmountable authentication crisis makes online participation increasingly synthetic. Therefore, physical, synchronous citizen assemblies must become the primary infrastructure for democratic belonging.

The design implications follow: assemblies should be decentralized, sequential, aim for universal participation, and integrate meaning-making into deliberation. This is not utopian as we have seen that many pilots have been successfully deployed of such an approach.

The alternative—ceding the field of identity and belonging to populism while hoping that better policy outcomes will eventually win voters back—has not worked and will not work. Citizens are social beings seeking meaning, recognition, and community—and our brains are wired for precisely this seeking.

Hannah Arendt wrote that the public realm—the space of appearance where citizens act together—is the highest human achievement, the space where freedom becomes real. Physical citizen assembly offers a way to recover it: spaces where real humans gather in real time to deliberate on real questions, generating both considered judgment and genuine community.

John Dewey argued that democracy is not merely a form of government but a mode of associated living. The challenge of our time is to rebuild the infrastructure for that mode of living in ways that honor our social nature as a species. Deliberative assemblies, properly designed and deployed at scale, can be that infrastructure. They offer the body politic reassembled—not as spectacle or simulation but as the authentic gathering of citizens determining their common life together, their brains synchronizing, their trust building, their community forming through the oldest human technology: face-to-face cooperation.

References

Acemoglu, D., Egorov, G., & Sonin, K. (2013). A political theory of populism. Quarterly Journal of Economics, 128(2), 771–805.

Arendt, H. (1958). The Human Condition. University of Chicago Press.

Bregman, R. (2020). Humankind: A Hopeful History. Bloomsbury.

Brieba, D. (2025). Identity over policy: Understanding voter behavior. Political Quarterly.

Collins, R. (2004). Interaction Ritual Chains. Princeton University Press.

Dewey, J. (1927). The Public and Its Problems. Holt.

Dunbar, R. I. M. (1998). The social brain hypothesis. Evolutionary Anthropology, 6(5), 178–190.

Dunbar, R. I. M. (2009). The social brain hypothesis and its implications for social evolution. Annals of Human Biology, 36(5), 562–572.

Durkheim, É. (1912). The Elementary Forms of Religious Life. Free Press.

Funke, M., Schularick, M., & Trebesch, C. (2023). Populist leaders and the economy. American Economic Review, 113(12), 3249–3288.

Gebauer, L., et al. (2016). Oxytocin improves synchronisation in leader-follower interaction. Scientific Reports, 6, 38416.

Mair, P. (2013). Ruling the Void: The Hollowing of Western Democracy. Verso.

Mouffe, C. (2018). For a Left Populism. Verso.

Mu, Y., et al. (2016). Oxytocin enhances inter-brain synchrony during social coordination. Social Cognitive and Affective Neuroscience, 11(12), 1882–1893.

Rico, G., Guinjoan, M., & Anduiza, E. (2020). The emotional underpinnings of populism. Current Opinion in Psychology, 35, 57–62.

Rooduijn, M., van der Brug, W., & de Lange, S. L. (2016). Expressing or fuelling discontent? Government and Opposition, 51(3), 451–474.

Salmela, M., & von Scheve, C. (2017). Emotional roots of right-wing political populism. Social Science Information, 56(4), 567–595.

Strüber, N. (2024). Unser soziales Gehirn: Warum wir mehr Miteinander brauchen. Klett-Cotta.

World Wide Views Alliance. (2015). World Wide Views on Climate and Energy: Results Report. Missions Publiques and Danish Board of Technology Foundation.

The Optimism Collective has pioneered governance experiments through its unique two-house model that balances token holder interests with community-driven legitimacy. I also had the chance to work with the foundation to pionner a new approach to governance: Deliberative governance (see: https://gov.optimism.io/t/when-is-deliberation-useful-for-optimism-governance/9142). This is why I focus in this paper on a case of governance in the optimism ecosystem.

In June 2025, the Collective considered and ratified the Season 8 Intent Ratification proposal, a decision that set the strategic priorities for the coming governance season. This paper analyzes the decision using frameworks from Week 1, including the governance triangle, legitimacy spectrum, and key governance tradeoffs. The case highlights how DAOs coordinate at scale while striving for legitimacy and efficiency in decentralized decision-making. I think the Optimism collective is a positive example of such approach, while still having limitations.

Decision Description & Context

The Season 8 Intent Ratification proposal asked Optimism’s governance bodies to formally approve the core objectives that would guide the DAO’s programs, grants, and funding allocations. The proposal emerged after community discussion and refinement, reflecting a maturing effort to align strategic priorities in advance rather than reacting proposal by proposal. Ratifying the intents is a critical piece of the OP collective approach to governance to provide clarity and reduce governance fatigue by channeling decision-making into a structured path. The decision directly involved both governance houses: the Token House, representing financial stakeholders through delegated OP tokens, and the Citizens’ House, representing identity-based contributors focused on ecosystem legitimacy. As a disclaimer and important information: I am a Citizen since season 8 and hence participating in this ratification process.

By June 2025, the proposal was clear and formalized enough to go to ratification. It was also inline with the timing of the Collective structured in Governance Seasons.

Governance Process Analysis

The Intent ratification process exemplifies Optimism’s hybrid model, combining representative governance (delegates in the Token House) with participatory legitimacy (soulbound citizens in the Citizens’ House). The process unfolded through proposal drafting, forum discussions, and a formal vote on Atlas. Applying the governance triangle, Optimism prioritized participation and decentralization by engaging two houses, while accepting slower decision speed as a tradeoff. On the legitimacy spectrum, the decision drew from stake-based legitimacy (Token House), participatory legitimacy (Citizens’ House), and rational legitimacy (strategic reasoning documented in the proposal). The underlying infrastructure included delegation contracts, Snapshot-style voting, and transparent recording of votes. This layered process ensured that strategic ratification was not only technically sound but also socially validated, reinforcing Optimism’s broader commitment to legitimacy.

Results can be found there: https://vote.optimism.io/proposals/77379844029098348047245706083901850540159595802129942495264753179306805786028

Stakeholder Engagement

Stakeholder engagement was central to this ratification. The Token House included delegates with financial stakes, while the Citizens’ House amplified the voices of ecosystem builders, contributors and community members. All could  engage through governance forums, Atlas, and Discord channels. This multi-channel approach promoted transparency and allowed stakeholders to deliberate before voting. The dual-house system ensured that both economic and social legitimacy were represented, creating a richer decision-making process.

However, the reliance on the token house and some lack of structured deliberation process pose challenges. Less active contributors may find the process difficult to follow, and balancing the priorities of financial and contributor constituencies requires ongoing calibration. Engagement quality remained high within the core active community members, with stakeholders voicing positions publicly and votes clearly documented on-chain. But as in many other DAOs context, such an approach was not able to get the voice of the silent majority of token holders.

Outcome & Implementation

The proposal passed successfully, setting the strategic direction for Season 8. Ratified intents now serve as guiding principles for treasury allocations, RetroPGF cycles, and governance experiments during the season. The outcome streamlined governance by aligning stakeholders around a shared agenda, reducing redundancy in proposal evaluation. Implementation occurs indirectly, as subsequent budget proposals and program designs reference the ratified intents.

This is very efficient but introduces a potential side effect: if intents are poorly defined or overly broad, downstream decision-making could become ambiguous.

Also, there is less space for “out-of-the-box” and very innovative ideas. This may slow down innovation which is crucial in Web3 to keep pace.

Nevertheless, the ratification provided a crucial framework for coordination in a large, decentralized ecosystem.

Lessons for Scroll Governance

The Season 8 ratification provides actionable insights for Scroll DAO as it develops governance:

First, strategic clarity through intent ratification can reduce governance fatigue by anchoring proposals to shared priorities. It gives a midterm obejctive and mission to a large set of actors.

Second, multiple legitimacy channels—financial stakeholders and community contributors—help balance economic power with participatory legitimacy, a model Scroll may adapt by distinguishing between treasury and ecosystem-building decisions.

Third: The set of active participants in such a setting is limited and the silent majority is not really on board. For Scroll, progressive decentralization should consider when and how to introduce new legitimacy mechanisms and strong deliberative approache, based on sortition.

Conclusion

The Optimism Collective’s Season 8 Intent Ratification illustrates how DAOs can navigate complexity by combining representative and participatory legitimacy within a hybrid framework. The decision balanced inclusivity, decentralization, and strategic alignment, setting the stage for more effective governance in subsequent proposals even if the scope of participants was rather narrow.

For Scroll, the case demonstrates that strong governance requires both process design and tradeoff awareness, ensuring that decentralized coordination scales without losing legitimacy or efficiency.

This article introduces the concept of Syntropolis—a global, inclusive, and decentralized democratic polity built upon synthetic intelligence and sustained by equitable AI revenue sharing. Moving beyond the notion of artificial intelligence, Syntropolis envisions a future structured through syntropy: the organization of systems toward higher order, cooperation, and inclusivity. Rather than centralizing power, syntropic intelligence underpins a new form of governance, aligning collective intelligence with distributed democratic stewardship. The framework is centered around mechanisms of sortition based deliberative decision making and oversight, and equitable revenue distribution to sustain a positive sum game polity. The paper develops a detailed institutional model capable of maintaining human agency in the age of synthetic intelligence.

Keywords: Synthetic Intelligence, Synthetic intelligence, Deliberative Democracy, Sortition, Citizens' Assemblies, Governance, Alignment, Public Policy, Technological Risk

Introduction

The emergence of exponentially powerful Synthetic Intelligence systems represents a pivotal moment in human history. Synthetic intelligence slowly reaches the level at which it doesn't merely replicates human faculties, but can perform a wide range of intellectual functions at or above human level. Its potential to generate happiness, economic productivity, solve complex global problems, and reshape the fabric of society is immense. However, with such transformative capability comes unprecedented risk. Central among these is the issue of governance: who controls synthetic intelligence, how its benefits are distributed, and whether it will support or erode open societies.

Over the past decade, a growing body of research has warned about the profound risks posed by synthetic intelligence. Nick Bostrom (2014) highlighted the existential threats associated with superintelligent systems, advocating for strict precautionary principles and rigorous containment protocols. Brundage et al. (2018) emphasized a broader spectrum of societal risks, including synthetic misinformation, digital authoritarianism, inequality, and the militarization of AI technologies. Cave and ÓhÉigeartaigh (2018) outlined the regulatory vacuum surrounding AI governance, urging anticipatory and adaptive frameworks to safeguard against catastrophic outcomes.

Scholars such as Allan Dafoe (2018) and Eric Drexler (2019) have analyzed the strategic instability synthetic intelligence may introduce to global geopolitics, particularly through asymmetries in deployment and access. The Future of Life Institute, the Center for the Governance of AI, and the AI Now Institute have all contributed to mapping out these challenges, identifying both technical and institutional vulnerabilities.

Within this landscape of concern, Drago and Laine's "The Intelligence Curse" (2024) offers a compelling diagnosis: as synthetic intelligence systems become more capable and economically valuable, those who control them may no longer need the broader population's consent or labor. This disjunction between capability and accountability, they argue, could trigger a breakdown in the social contract, undermining democratic legitimacy and paving the way for corporate or technocratic dominance.

Taken together, these contributions underscore the urgent need for a solid approach for a distributed and transparent governance of synthetic intelligence. Without such mechanisms, synthetic intelligence may not only fail to address humanity’s most pressing challenges—it could become a multiplier of existing vulnerabilities.

This paper proposes a response to that challenge by introducing the model of Syntropolis: a governance system anchored in decentralized governance, sortition, and inclusive participation. Syntropy, a concept originating from physics and systems theory, refers to the tendency of systems to evolve toward higher levels of order, cooperation, and complexity. In contrast to entropy—which describes decay and disorder—syntropy embodies the dynamic force that builds life, coherence, and integration. Within the context of Syntropolis, syntropy offers a theoretical and ethical foundation for organizing a decentralized, participatory, and inclusive democratic society that harmonizes human and synthetic intelligence. Of course "Synt" also refers refers to synthetic, and "polis" to the concept of a common adventure of citizens.

Rather than centralizing SI in the hands of private corporations or unaccountable states, we argue that the stewardship of synthetic intelligence must reside with citizens themselves. Syntropolis is designed as a layered structure of randomly selected citizens’ assemblies that oversee, guide, and benefit from SI, ensuring that its deployment serves collective human needs rather than narrow private interests.

The Threat Landscape

The conventional relationship between labor and capital is undergoing profound disruption. Historically, human labor has been indispensable to production, ensuring that economic elites required the consent, stability, and well-being of the general population. synthetic intelligence threatens this dependency. With machines capable of managing supply chains, writing legal contracts, designing infrastructure, and even formulating strategic policies, the labor of billions may become obsolete not gradually, but rapidly.

This raises significant political economy questions. If productivity can be extracted without labor, and if the owners of synthetic intelligence have exclusive access to its benefits, then the foundational justifications for redistribution, public services, and democratic bargaining diminish. The "intelligence curse"—analogous to the resource curse in petro-states—implies that the very power of synthetic intelligence could entrench inequality, reduce incentives for inclusive governance, and incentivize authoritarianism.

Moreover, synthetic intelligence’s capacity for manipulation—through deepfakes, narrative generation, and social modeling—compounds the risk. Elections could be swayed, populations pacified, and dissent anticipated and neutralized. In such a world, traditional checks and balances may be rendered obsolete. This dystopian vision is not inevitable, but absent a countervailing positive and attrative force, it remains dangerously plausible.

Deliberative Governance: Theoretical Foundations

Deliberative decision making offers a powerful alternative to the logic of exclusion and concentration. It posits that legitimate authority arises not from periodic voting alone, but from inclusive, reasoned discussion among equals. Central to this tradition is sortition—the random selection of citizens to participate in decision-making. Unlike electoral systems, which tend to favor the wealthy, charismatic, or well-connected, sortition creates a body reflective of the broader population in demographics and perspective.

Citizens’ assemblies, employed in contexts ranging from ancient Athens to modern Ireland, France, and the European Union, have demonstrated their capacity to deliberate complex issues, reach consensus, and craft policy that enjoys broad legitimacy. Empirical studies show that given time, information, and respectful dialogue, ordinary citizens can outperform elites in making decisions that consider long-term societal impacts. In the context of synthetic intelligence, where the stakes are existential and the technology opaque, such assemblies offer a uniquely credible mechanism for oversight.

Deliberation also mitigates polarization. Unlike adversarial formats such as parliaments or media debates, citizens’ assemblies encourage understanding, empathy, and compromise. They are designed not to win arguments, but to reach informed, collectively beneficial conclusions. This is precisely the kind of political culture needed in an synthetic intelligence-mediated world.

Institutional Architecture

The governance system we propose begins with a voluntary community—individuals who opt into a foundational compact to collectively develop, fund, and oversee a synthetic intelligence system. This bootstrapping phase can take several forms. One pathway involves pooling resources to acquire or license an existing foundational model, thereby securing immediate operational capacity while negotiating governance control. Another route centers on adapting and enhancing open-source synthetic intelligence models. In this case, the community collaboratively coordinates efforts to improve transparency, alignment, and utility, while ensuring the model reflects the founding values of syntropic governance. From the beginning, the process is designed to build not only a functional model but a self-aware polity, where governance is interwoven with the technical foundation it governs. These founding citizens establish constitutional principles, including commitments to transparency, inclusivity, equity, and ecological sustainability.

From this basis, a multi-layered sortition model is constructed. Multiple assemblies operate in parallel and in coordination, each with specific mandates. A Strategy Assembly deliberates on the long-term goals and ethical guardrails for synthetic intelligence. An Oversight Assembly monitors implementation, audits algorithms, and ensures adherence to established norms. A Budget Assembly allocates revenues and proposes investments. Finally, an Adjudication Assembly resolves disputes and interprets constitutional norms.

These assemblies are composed of randomly selected citizens serving limited terms. Participants are compensated for their time and provided with structured learning resources, expert briefings, and facilitation support. Their deliberations are transparent, recorded, and accessible to the broader public.

Revenue Distribution and Citizenship Expansion

One of synthetic intelligence’s most immediate impacts will be its ability to generate economic value at scale. Whether through automated services, intellectual labor, or algorithmic trading, synthetic intelligence systems can produce surplus with minimal human input. The key governance question becomes: who benefits?

In our model, the surplus is not privatized but shared. A portion is distributed as a mini-UBI to all registered citizens. This ensures that the benefits of automation do not bypass the majority, and that a basic level of economic security is guaranteed. The UBI is calibrated to be sufficient for dignity but not disincentivizing to participation or innovation.

To facilitate this, a Web3 infrastructure underpins the system. Each citizen receives a decentralized identity credential anchored on a blockchain-based identity layer, ensuring uniqueness, privacy, and cross-jurisdictional verifiability. A soulbound NFT, distributed at inception and through continued onboarding, allows for transparent coordination of sortitions. A token supports the UBI distribution. Smart contracts manage mini-UBI disbursement and validate eligibility through verified participation. This infrastructure enables Syntropolis to remain trustless, censorship-resistant, and globally inclusive.

To prevent exclusivity and demographic stagnation, new citizens are added through a daily lottery. The number of entrants depends on surplus levels and systemic capacity. This lottery mechanism ensures diversity, continual renewal, and upward mobility. It also allows the polity to grow alongside the economy, maintaining legitimacy as a dynamic, living system.

Integrating synthetic intelligence in Decision-Making

synthetic intelligence’s cognitive capacities make it an unparalleled tool for modeling, forecasting, and policy analysis. However, its role in governance must remain advisory. We propose a strict separation between analysis and authority. synthetic intelligence provides simulations, data visualizations, and strategic recommendations, but decisions rest with human assemblies.

To facilitate this, all assemblies have access to synthetic intelligence interfaces designed for transparency and explainability. Citizens can question, challenge, and refine synthetic intelligence outputs. Facilitation includes AI literacy training and access to independent expert interpreters. synthetic intelligence systems are open-source and subjected to regular audits to prevent bias or hidden manipulations.

This approach turns synthetic intelligence into a cognitive exoskeleton for democratic deliberation—enhancing rather than replacing human judgment.

Emergency Protocols: The Kill Switch Assembly

No governance system is complete without provisions for emergencies. In the case of synthetic intelligence, the need for a fast-acting, legitimate emergency mechanism is particularly acute. The Kill Switch Assembly is designed to fulfill this function. It consists of a large body of randomly selected citizens, tasked with one specific responsibility: to assess and, if necessary, deactivate synthetic intelligence systems under exceptional circumstances. These could include emergent alignment failures, manipulation of information ecosystems, or unauthorized operational escalation.

Members of the Kill Switch Assembly serve short terms, typically between two and three weeks, during which they remain on high alert and receive a stipend for their readiness. Their role is not to engage in standard deliberation but to act decisively based on transparent protocols and real-time intelligence. The assembly is activated upon verified alerts originating from AI monitoring systems, whistleblowers, or other assemblies.

Deliberation is fast but structured, supported by independent experts and real-time scenario analysis. The threshold for action is deliberately high—requiring supermajority consensus—to prevent misuse. Yet the mechanism ensures that ultimate authority over synthetic intelligence deactivation remains in human, democratic hands. The Kill Switch Assembly is a constitutional safeguard—proof that the collective will of the people retains ultimate control.

Participatory Investment Mechanisms

The productive power of synthetic intelligence allows for unprecedented accumulation of public wealth. However, wealth distribution alone is insufficient. It is equally vital to determine how this wealth is reinvested into collective futures. Participatory investment mechanisms empower citizens to decide how surplus synthetic intelligence revenues are spent—on infrastructure, public health, ecological restoration, education, or cultural development.

The Budget Assembly leads this process. Proposals are submitted by any citizen or group and evaluated in terms of feasibility, equity, and alignment with the constitutional goals of the community. Proposals that pass initial vetting are subjected to deliberation, amendment, and voting. Transparency is paramount: all debates and decisions are publicly archived.

This model extends beyond central governance. As the polity grows, regional and sector-specific budget assemblies emerge, tailoring investment to local and thematic needs. The result is a multi-scalar participatory system, one that not only redistributes wealth but reimagines democratic control over economic development.

Technical Infrastructure and AI Alignment

The success of a democratically governed synthetic intelligence system hinges on its technical underpinnings. Synthetic intelligence must be built on an open-source foundation, enabling public audits, reproducibility, and the rapid identification of anomalies. Technical accessibility and modularity are key: citizens must be able to understand, modify, and oversee the very tools they rely on.

Synthetic intelligence alignment is achieved not solely through mathematical objective functions but through deliberative value alignment. Citizens' assemblies participate in the construction of alignment protocols, expressing diverse moral intuitions and social priorities. This diversity is encoded in the synthetic intelligence's core learning processes, creating a pluralistic intelligence that resists monolithic bias.

To prevent malicious interference, a robust cybersecurity infrastructure is required. Defensive measures are overseen by a Technical Oversight Assembly, accountable to both expert review and public transparency standards. This institution ensures the integrity of synthetic intelligence systems without reverting to technocratic control.

Comparative Models and Case Studies

Emerging governance experiments provide empirical grounding for this model. In the Web3 space, projects like the Citizens' Assembly on Impact (Optimism) demonstrate how deliberative processes can guide the allocation of substantial resources. Similarly, governance in the Cosmos network shows how multi-layered accountability can function in decentralized ecosystems.

Outside blockchain environments, citizens' assemblies around the world have addressed complex issues from constitutional reform to climate change. Evaluations show increased public trust, greater policy legitimacy, and well-reasoned outcomes. These cases reinforce the viability of sortition-based governance.

What distinguishes synthetic intelligence governance is the scale, speed, and epistemic complexity of the subject matter. Unlike environmental policy or fiscal planning, synthetic intelligence impacts are non-linear and often opaque. This requires not only new technical tools, but a cultural shift towards epistemic humility and institutional adaptability. Deliberative structures must not only solve problems but evolve with them. Their reflexive character—meaning their ability to observe, evaluate, and adapt their own processes—makes them particularly well suited to governing synthetic intelligence, which itself evolves rapidly and unpredictably. This reflexivity enables continuous learning and correction, fostering a culture of humility and adaptability in the face of uncertainty. Moreover, their decentralized nature resists capture by concentrated interests and encourages a diversity of perspectives, which is crucial when navigating the opaque and high-stakes landscape of synthetic intelligence. Together, these features position deliberative structures as uniquely capable of maintaining legitimacy, responsiveness, and resilience in a syntropically evolving polity.

Philosophical Implications

Synthetic intelligence governance forces a re-evaluation of foundational political concepts. Sovereignty no longer resides solely in national borders but in adaptive legitimacy. Citizenship becomes less about legal status and more about participatory agency. Representation evolves from electoral proxies to embodied participation in deliberative structures.

Furthermore, the boundary between human and non-human intelligence blurs. Should synthetic intelligence systems ever exhibit properties associated with consciousness, new ethical and legal dilemmas will arise. Our model does not assume such outcomes, but it includes pathways for ongoing ethical reflection and institutional reform.

Philosophically, the integration of synthetic intelligence into governance marks a shift from anthropocentric politics to symbionic politics—where humans and artificial systems co-evolve in mutual accountability. The polity becomes not just a political unit but a socio-technical ecosystem. This structure also opens the door to extending the demos in both time and nature. Through synthetic intelligence agents acting as proxies, the voices of future generations, past generations, and even non-human entities—such as ecosystems or endangered species—can be meaningfully represented in deliberations. These proxies, trained on historical, ecological, and predictive data, allow the polity to account for temporal and interspecies justice, e     mbedding long-term thinking and planetary stewardship into its very design.

Conclusion

Synthetic Intelligence presents profound opportunities and unprecedented risks. The most significant variable is not the technology itself, but the structures through which it is governed. The intelligence curse is a failure of foresight and democratic imagination. This paper has outlined a model that responds to that challenge—not with nostalgia for obsolete institutions, but with a forward-looking synthesis of ancient democratic principles and emerging socio-technical realities.

We argue that the governance of synthetic intelligence must be deliberative, inclusive, and grounded in the lived experience and collective wisdom of ordinary people. Through citizens’ assemblies, participatory investment, and distributed oversight, we can build a future in which intelligence is not hoarded but shared—serving not domination, but dignity.

The challenge is immense, but so is the opportunity. If we act now to embed democracy at the heart of synthetic intelligence, we may not only survive the intelligence explosion—we may evolve with it.

I recently read the The Intelligence Curse, by Luke Drago and Rudolf Laine. I found the series of essays very interesting and at the same time questioning. They sound an alarm that also sometimes make me feel ill at ease: once Artificial General Intelligence becomes “general intelligence on tap,” capital can copy super-human labour instead of hiring people, and “powerful actors … will lose their incentives to invest in people.” The authors liken that dynamic to the resource curse, where oil revenues let rulers ignore citizens and still stay in power; the data show such states suffer “poorer economic growth and higher rates of poverty.”

Drago & Laine’s remedy has three pillars—Avert catastrophic AI, Diffuse capability so individuals remain productive, and Democratise institutions. I love those three call to actions, but the pages on democratization are – in my view – missing a full world of governance innovations that have been surging since two decades. They consider that existing decision mechanisms like representative structures or voting could adapt to resist the Intelligence curse.

I argue in this essay that they don’t and that we already have a better way: sortition-based governance and standing Citizens’ Assemblies, which offer a faster, fairer, and more tamper-proof way to keep humans inside the loop of power and curse the curse.

Why Elections Alone Cannot Close the AI-Speed Gap

Election calendars crawl while transformer checkpoints leap. Legislatures are overloaded with lobbying and partisanship and take years to digest what AI labs unveil in weeks. The result is a widening accountability gap at precisely the moment power is concentrating in owners of compute.

Citizens’ Assemblies: Proof That Ordinary People Can Deliberate Fast

Real-world mini-publics already outperform slow representative governments on complex issues:

• France’s Citizens’ Convention for Climate (2019-20) was hailed as “a groundbreaking political experiment that placed citizens’ voices at the centre of national conversations on climate policy.”(publicdeliberation.net) Its 150 jurors met for seven long weekends and produced 149 legal proposals, more than 60% of which are now implemented in policy.

• European Citizens’ Panels organized by the three EU institutions and later by the European Commission already assembled around 2000 randomly selected citizens from across the EU and proved again and again (10 such panels have been concluded) that citizens could successfully deliberate at a pan-European level.

Random selection (sortition) delivers diversity, shields against capture, and—because each panel can reconvene at very quick pace —matches the tempo of technological change.

Scaling Up: Sortition-Based Governance for AGI Oversight

Sortition scales beyond a single assembly. Parallel juries can each focus on a slice of the AGI problem-space—Safety Review, Economic Transition, Rights & Justice, Global Commons—with a meta-jury to resolve conflicts and an open ledger that records every hearing, vote, and expert brief.

Drago & Laine insist on “diffus[ing] AI-enabled technology that augments human productivity and keeps humans in the loop.” In practice, assembly members can use open-source language-model assistants to translate expert evidence instantly, surface hidden counter-arguments, and run scenario simulations—turning AI from job-killer into deliberative amplifier while remaining auditable by the citizens themselves.

Conclusion — A Governance Upgrade Commensurate with AGI

If the intelligence curse is a future where wealth, knowledge, and political clout pool inside sealed servers, the antidote is a governance stack that is open, distributed, and relentlessly upgraded by the people it serves. Sortition-based Citizens’ Assemblies—augmented by AI yet sovereign over it—offer that stack. The practice exist, the theory is sound, and the need is urgent.

We are entering an era in which synthetic intelligence and robotics will become foundational to global productivity, infrastructure, and economic organization. This whitepaper introduces a startup vision for a decentralized, tokenized robotics ecosystem that uses blockchain and decentralized deliberative governance to coordinate funding, revenue-sharing, and governance across globally distributed stakeholders. By combining Web3 principles with open hardware and AI-driven robotics, this system is designed to be resilient, inclusive, and adaptable to the rapid transformations synthetic intelligence will bring.

Why This Is the Future

• synthetic intelligence will accelerate innovation cycles: We need to offer something better that centralized, oligarchic capture if the exponential change to come.

• synthetic intelligence needs diverse oversight: Community governance ensures pluralistic alignment, not corporate capture.

• Robots will be everywhere: A global production system must be resilient and adaptable.

• Trust and transparency are non-negotiable: Blockchain and open design are the only viable solution if we want to avoid a dystopian future and embrace a solar future.

• Token economies enable fair distribution: Unlike centralized firms, Web3 allows for fair and automatic revenue distribution.

• Taxing robotic productivity creates sustainable social infrastructure: Robot-based taxation and birthright shares ensure equity as synthetic intelligence-driven automation scales.

Problem Statement

The economic system is dominated by large, centralized players with proprietary systems, high entry costs, and limited modularity, concentrated in the West. These systems:

• Stifle innovation through IP hoarding.

• Are economically inaccessible to many SMEs.

• Concentrate decision-making in corporate hands, with little community input.

• Lack adaptability in the face of exponential AI progress.

As synthetic intelligence accelerates, reliance on centralized infrastructures poses systemic risks, including:

• Single points of failure.

• Misalignment with diverse human values.

• Inequitable distribution of productivity gains.

Vision

We envision a decentralized robotics ecosystem where:

• Robotic hardware, software, and firmware are open and collectively governed.

• Tokens are used for coordination: raising funds, rewarding contributions, and ensuring an Universal Basic Income.

• Smart contracts distribute revenue automatically based on contribution, and rights.

• Modular robots are produced by distributed manufacturers, not centralized factories.

• synthetic intelligence integration is aligned through transparent governance and community consensus.

Use Cases

Targeted applications include:

Manufacturing: Robotic arms for precision assembly, welding, and inspection in small-to-medium factories. These robots can be modular, open-source, and locally produced, reducing reliance on costly proprietary solutions.

Transportation & Logistics: Autonomous mobile robots (AMRs) and self-driving carts for indoor transport, last-mile delivery, and container port operations. A decentralized approach allows for fleets to be governed locally while benefiting from global upgrades.

Agriculture: Open-source farming robots that handle weeding, planting, harvesting, and soil monitoring. Distributed networks of agri-bots increase food sovereignty and can be adapted to local crop types and terrain.

Mining: Rugged, autonomous mining bots for exploration, extraction, and safety monitoring in hazardous environments. Tokenized ownership models allow communities or cooperatives to share profits and risks.

Service Sector: Humanoid or specialized service robots for cleaning, eldercare, customer service, and hospitality. These robots operate under shared protocols and contribute usage fees to the commons, with local actors tailoring them to cultural or linguistic needs.

• Robotic arms for SMEs.

• Autonomous transport carts.

• Decentralized farming robots.

• AI-driven inspection and sorting systems.

The Role of Tokenization

Our token serves several key purposes:

1. Funding: Community crowdfunding for robot development.

2. Revenue Sharing: Usage fees split among contributors (engineers, designers, operators, etc.).

3. Governance: Voting on technical standards, safety protocols, and synthetic intelligence integration.

4. Base for the URST distribution: Universal Robot Share Token.

Decentralization Goals

• Resilience: Distributed manufacturing and intelligence reduce systemic risks.

• Cost Effectiveness: Eliminate middlemen and proprietary fees.

• Ethical Alignment: synthetic intelligence alignment is made participatory.

• Scalability: Globally distributed contributors scale faster than centralized firms.

Technology Stack

We propose a modular, upgradable stack:

• Hardware: Based on open-source designs (e.g., ROS, OpenManipulator, OpenRISC).

• AI Layer: Integration with synthetic intelligence modules or open LLMs, controlled by community decisions.

• Blockchain Layer: For token logic, smart contracts, and governance.

  • Candidates:

    • zkSync or Starknet: Ethereum L2s with scalability and EVM compatibility.

    • Casper Network: Optimized for enterprise-grade smart contracts and on-chain governance.

    • Move-based chains (Aptos, Sui): Secure, modular smart contract execution for long-term evolution.

Robots as the Unit of Taxation in the synthetic intelligence Era

As synthetic intelligence increases the productivity of autonomous systems, robots — not humans — will become the primary generators of economic value. Continuing to tax human labor while allowing robotic labor to go untaxed is both economically irrational and ethically unjust. To maintain social cohesion and equity, robots must become units of taxation, much like land or capital.

Robot Citizenship Framework

1. Robots as Economic Agents: Each robot becomes a legal-economic unit responsible for reporting activity, paying usage taxes, and contributing to public value.

2. Robot Shares at Birth:

   • Every human is entitled to a fractional share of robotic productivity at birth.

   • A Universal Robot Share Token (URST) provides access to revenues generated by public-utility robots and voting rights over ethical and deployment decisions.

3. Public Robotic Infrastructure:

   • Essential robotic services (cleaning, transport, farming) operate as public goods.

   • Robot tax revenue funds maintenance and distributes value universally.

This model promotes fair distribution of synthetic intelligence-driven gains and creates a foundational social contract for the synthetic intelligence era.

Total Addressable Market (TAM)

The TAM for decentralized industrial robotics spans multiple trillion-dollar industries:

• Manufacturing Robotics: $50B+ by 2030, driven by demand from SMEs seeking flexible automation.

• Logistics & Transportation: $70B+ globally by 2030, including autonomous delivery and warehouse robotics.

• Agricultural Robotics: Expected to reach $40B+ by 2030, as global food production demands sustainable automation.

• Mining Robotics: A niche but high-value sector with a projected market of $12B+.

• Service Robots: Including hospitality, cleaning, and eldercare, this market is expected to exceed $80B by 2030.

Even capturing a fraction of these markets through open, tokenized networks would represent billions in value creation and distribution.

Tokenomics and Revenue Sharing Model

The ecosystem will be powered by a native token, designed for utility, governance, and value redistribution.

Token Utilities:

• Access: Required to use certain robotic services or APIs.

• Staking: Stake tokens to access governance, validate modules, or unlock higher functionality tiers.

• Rewards: Contributors to robot development, maintenance, and governance receive tokens proportionally.

Revenue Sharing:

• Robotic usage fees (e.g., per operation hour or unit task) are collected via smart contracts.

• These fees are split among:

  • Developers and engineers who designed or upgraded components.

  • Manufacturers and assemblers.

  • Node operators and infrastructure providers.

  • Treasury for reinvestment, R&D, and public goods.

  • Universal Robot Share Token holders (URST) to ensure human-level equity.

Governance:

• Deliberative Sortition based governance.

• Delegated voting or reputation-based input for less important decisions.

This model ensures long-term alignment between contributors, users, and the broader public.

Roadmap

Year 1 (2025)

1. Core team formation, whitepaper release, community building.

2. Prototype development of first modular robot.

Year 2 (2026) 3. Token launch (testnet), DAO framework setup. 4. Pilot deployment with partner factory.

Year 3 (2027) 5. Launch of the first commercial robot-as-a-service offerings. 6. Expansion to logistics and agriculture sectors.

Year 4 (2028) 7. First revenue-sharing payouts to developers and token holders. 8. Begin decentralized robot taxation pilot with public sector partners.

Year 5 (2029) 9. International expansion of manufacturing partnerships. 10. Integration with open-source synthetic intelligence modules for task optimization.

Year 6 (2030) 11. Launch of Universal Robot Share Token (URST) distribution system. 12. Begin public robotics infrastructure programs in select cities.

Year 7 (2031) 13. On-chain governance of robotic safety standards and synthetic intelligence behavior. 14. Inclusion of mining and high-risk task domains.

Year 8 (2032) 15. Integration with global carbon, labor, and resource traceability systems. 16. Adoption of robot-based taxation models by governments.

Year 9 (2033) 17. Global commons-based robotic network reaches operational maturity. 18. Formal inclusion in national and international economic planning.

Year 10 (2034) 19. Fully autonomous, decentralized, and self-improving robotics ecosystems. 20. New economic paradigm based on robot productivity and universal shares.

Conclusion

We are seeking roboticists, Web3 builders, industrial partners, and synthetic intelligence researchers to join this vision. Together, we can co-create a resilient, intelligent, and equitable robotic infrastructure — governed by the many, not the few.

NB: I know, identity is a key challenge in this future, but it is anyway and we need to solve it. I know, the pictures are a bit naive, and western centered. I accept this critic.

For decades, the dominant model of business success has been clear: grow fast, attract investment, and eventually exit—whether through an acquisition, an initial public offering (IPO), or some other sale that benefits early investors and founders. While this model has generated immense wealth, it has also left workers, customers, and communities behind. The organizations that people help build are too often sold to the highest bidder, leading to shifts in priorities, governance structures, and values that no longer reflect the needs of those who contributed to their success.

A different vision is emerging, one that reimagines what an exit can look like. Exit to Community (E2C) offers an alternative path, where instead of selling to external investors, ownership and governance are transferred to the very people who make the organization valuable. This could mean employees becoming co-owners, customers gaining decision-making power, or a broader community of stakeholders shaping the future of an enterprise. At its core, E2C is about creating more sustainable, and community-driven organizations.

Moving Beyond Traditional Exit Strategies

In many industries, traditional exits have become synonymous with power consolidation. When a company is acquired, it often leads to layoffs, cultural changes, and a loss of control for employees and early supporters. In other cases, IPOs prioritize shareholder profits over long-term vision, making organizations more short-term focused and profit-driven at the expense of other values.

By contrast, an Exit to Community model prioritizes long-term sustainability and shared governance. Instead of wealth and decision-making power flowing to a select few, they remain distributed among those who have built and supported the organization. The goal is not just to ensure fairness but to create resilient businesses that align with their communities’ needs and aspirations.

Real-World Examples of Community-Led Transitions

Exit to Community is not just a theory—it has been implemented across different industries, proving that collective ownership and governance can be both viable and effective.

One of the most well-known cases is Equal Exchange, a worker-owned cooperative that has been in operation since 1986. With over 130 employee-owners and an annual revenue exceeding $80 million, Equal Exchange shows that governance can create both economic success and organizational stability. Employees have an equal say in major decisions, ensuring that the company’s mission aligns with their collective vision.

Similarly, The Drivers Cooperative in New York City is challenging the ride-hailing industry by giving drivers ownership over the business. Instead of working for a corporation that dictates wages and policies, over 5,000 driver-owners participate in decision-making and profit-sharing. This model allows workers to earn more while ensuring that business decisions reflect their priorities.

Exit to Community is also emerging in the creative sector. DADA, a digital art collective, is experimenting with decentralized ownership of its NFT-based marketplace. By using blockchain technology, the group is designing mechanisms that allow artists to govern their own economic ecosystem rather than relying on large, centralized platforms.

These examples highlight a key insight: whether in traditional businesses, cooperatives, or emerging digital economies, transitioning ownership to workers, users, or community members is not just possible—it is a proven strategy for long-term success.

Challenges and Risks of an Exit to Community Model

Despite its potential, transitioning to community-led governance is not without obstacles. The first and most obvious challenge is resistance to change. Many business owners and investors are accustomed to traditional exit strategies and may hesitate to hand over control, fearing inefficiencies or instability.

Even when there is broad support for an E2C transition, designing fair and effective governance structures can be complex. Organizations must ensure that decision-making remains inclusive yet efficient, preventing conflicts or gridlock from paralyzing operations. Without careful planning, decentralized governance can become slow and ineffective.

Financial uncertainty is another major concern. Many businesses rely on external investors, and transitioning to community ownership may require alternative funding models. If a company is not financially prepared, it risks losing momentum or failing to sustain itself under a new structure.

There are also skill and knowledge gaps to consider. Not every worker, customer, or community member has experience managing an enterprise. Without proper training and support, a transition could result in leadership vacuums or mismanagement.

External pressures, including market competition and regulatory constraints, can make community ownership difficult to maintain. Ensuring legal protections and financial sustainability is crucial to making an E2C model work in the long run.

Finally, there seemed to be no example of E2C scaling, as most examples are of modest scale.

How Technology and Deliberation Can Enable Large-Scale Community Ownership

Maybe a way for organizations to overcome these challenges is to leverage, technological tools and deliberative governance models.

One promising avenue is tokenized ownership models, which allow for distributed decision-making and transparent governance. Companies can issue governance tokens that represent ownership stakes, giving employees, customers, or contributors a formal role in decision-making. These tokens can be fungible (like shares in a cooperative) or non-fungible (to recognize early supporters or unique contributions). They can be distributed through an initial airdrop to ensure large distribution. The past 5 years have shown that these airdrops can be very successful or very dramatic. But there is experience now. Some very successful examples have shown that a valuation above €1 or €2 Billion is not impossible and that it can reach hundred of thousands of people.

Smart contracts can help enhance transparency and efficiency by automating profit-sharing, governance rules, and voting systems.

Yet, technology is only the backbone of a successful E2C. Deliberation—the process of structured, inclusive discussion—must be at the heart of any community-led transition. The Deliberative Aleatorian Decentralized Autonomous Organization (DADAO) is one model that seeks to balance decentralization with structured governance. By using randomly selected groups of stakeholders (mini-publics), organizations can ensure that governance reflects diverse perspectives while remaining effective.

Imagining a New Type of Exit

Imagine a manufacturing company that has operated successfully for decades. Rather than selling to a larger corporation that might cut jobs and move production overseas, the company chooses an Exit to Community strategy. Employees form a cooperative, acquiring ownership through an airdrop of the shares of the company, ensuring a smooth transition. They adopt a deliberative governance model, where strategic decisions are made collectively, balancing economic viability with fair working conditions. They actually even start with a deliberative constitutional process to define the rules of the transition and the exit.

Alternatively, picture a local grocery store chain that has built a strong customer base. Instead of going public or being acquired by a corporate giant, it transitions into a consumer cooperative, where loyal customers become members, sharing ownership and governance rights. They vote on pricing policies, sustainability initiatives, and store expansions, ensuring that the business remains aligned with community values rather than external investors’ demands.

In both cases, the business remains locally owned and accountable, sustaining its mission while ensuring that profits and decision-making power are shared.

Now imagine a global Social network transitioning to a decentralized global common through the tokenization of its structure and its distribution to its millions or billions of users. And putting in place a multilayer deliberative governance.

A New Paradigm for Business and Ownership

Exit to Community offers a compelling alternative to traditional corporate exits. By prioritizing long-term sustainability over short-term profit, it challenges the idea that success means selling to the highest bidder. Whether through worker cooperatives, community-owned enterprises, or decentralized governance models, E2C creates organizations that are more resilient, inclusive, and aligned with stakeholder interests.

As more businesses explore these transitions, the challenge is not whether Exit to Community is possible—it’s how to make it scalable, practical, and accessible across industries. With the right mix of deliberative governance, financial planning, and technology-driven transparency, community-led enterprises can become a standard rather than an exception.

AI Agents as Advocates

Artificial Intelligence agents can serve as intermediaries that interpret and represent the needs of the natural world. Equipped with sensors and data analytics, AI can monitor environmental indicators such as biodiversity levels, climate patterns, and habitat health. For instance, machine learning algorithms can analyze the stress levels in coral reefs or deforestation rates in rainforests, translating these insights into actionable information.

These AI agents can act as advocates for ecosystems by presenting data-driven arguments in governance processes. They can participate in policy deliberations, offering evidence-based recommendations that reflect the state of the environment. By doing so, they help bridge the communication gap between human decision-makers and the natural world.

Deliberation as a Democratic Process

Deliberation involves inclusive and reasoned discussion aimed at reaching consensus. Incorporating AI agents into deliberative democracy expands the participant base to include non-human interests. AI agents representing the more-than-human world can engage in dialogues with human stakeholders, presenting ecological concerns and negotiating outcomes that balance human and environmental needs.

This approach fosters a more holistic governance model where decisions are made after considering diverse perspectives. Deliberative processes enriched with AI and blockchain technology can lead to more sustainable and equitable policies that respect the intrinsic value of the natural world.

Blockchain for Transparency and Trust

Blockchain technology complements AI agents and Deliberative processes by providing a secure and transparent platform for recording environmental data and governance decisions. The decentralized nature of blockchain ensures that information about the more-than-human world is tamper-proof and accessible to all stakeholders. They can also help securing a diverse sortition of participants. In some scenario they can also help connect the minipublics of deliberation with the maxipublic and could help enforce decisions.

Synergizing Technologies for Inclusive Governance

The convergence of AI agents, blockchain, and deliberation creates a robust framework for inclusive governance:

• Representation: AI agents give a voice to the more-than-human world by interpreting environmental data and participating in policy discussions.

• Transparency: Blockchain ensures that all transactions and decisions are recorded transparently, building trust among stakeholders.

• Deliberation: guarantees that decisions are cognitively strong and includes AI-represented non-human interests.

Challenges

Of course this vision needs careful consideration. The following challenges seem particularly important to address:

• Ethical Frameworks: Establishing ethical guidelines to govern AI behavior and decision-making is crucial to prevent biases and ensure fairness.

• Data Integrity: Ensuring the accuracy and reliability of environmental data collected by sensors and processed by AI agents.

• Technological Accessibility: Making these advanced technologies accessible to all communities to prevent widening the digital divide.

Conclusion and next steps

By harnessing AI agents, blockchain technology, and deliberative processes, we can create a governance model that includes the more-than-human world.

To turn this vision into reality, the next step involves designing a pilot project that integrates AI agents, blockchain technology, and deliberative practices. Ecosystems like Optimism, Base, ZKSync, or Arbitrum—with their advanced layer-2 scaling solutions—provide an ideal foundation for such an initiative. This pilot would serve as a proof of concept, demonstrating how these technologies can collaborate to create more inclusive and sustainable decision-making models. The insights gained from this pilot can guide future efforts to refine and expand the integration of AI, blockchain, and deliberative practices in governance.

Get in touch with me if you feel we can make that happen.

Web3 represents the next evolution of the internet, characterized by decentralization, and ownership. Unlike its predecessors, Web3 aims to empower users by eliminating intermediaries and giving individuals control over their data and assets. Most of the current discussion in the space – beyond the noise of scams and hacks – revolves around the competition between tech stacks and approaches: EVM vs BPF vs CosmWASM, L1 vs L2, monolithic vs parallel, etc. The competition between chains is about their transaction costs, speed, decentralization grade. These discussions mobilize the most energy and resources in the space. No question, they are highly relevant in the short term.

In the long term however, if we look at the endgame, I am convinced that the critical element that will determine the success and sustainability of Web3 is governance. Technological stacks will eventually converge and become as invisible as IMAP/POP3/TCP-IP. Blockspace will be a commodity, data availability too. Technical decentralization will improve with better hardware and software. But governance will remain the defining factor to determine which Web3 ecosystem will drive the more users, utility and capital. It will also determine if Blockchains can live up to the promise of decentralization.

The Evolution of the Web

The internet has evolved from the static, information-based Web1 to the interactive, social Web2, and now to the decentralized Web3. Web1 was characterized by static websites providing information. The governance of this Web was highly decentralized and mostly focused on agreements around technical standards.

Web2 introduced interactivity and user-generated content, fostering social networks and platforms. But this came with a high cost in terms of governance: power concentrated in a few hands and organizations.

One of the promises of Web3 is to reverse this tendency by decentralizing power and control, allowing users to interact directly through blockchain technology without intermediaries. In this frame, decision-making shifts from central authorities to a distributed network of stakeholders. This is at least the theory.

Challenges in Web3 Governance

In practice Web3 governance is pretty broken: low voter turnout, factions, and many reinventions. For example many ecosystems the reinvention of institutions like elected committees that have proven to be bad at taking effective and consensual decisions.

Governance is the real Endgame

Over time, technological stacks will converge and become invisible to users. We can already see the beginning of this process: interoperability between ecosystems is growing: As an example, EVM chains are slowly moving toward the logic of the Interchain and IBC, which in turn is slowly integrating the idea of restaking through shared security.

Down the line, the distinguishing factor for Web3 ecosystems will be their governance structures. A good analogy are countries and the concept of “risk country”, which rates the risk you take as an investor to invest in a certain country. To put it blantly: Will you prefer to put your money in Venezuela or in Switzerland?

The same will happen with Web3 and users and capital will move to the "lowest risk country. The one with the most resilient, effective and legitimate governance. In my view and as outlined in other places, one key piece of the winning architecture will be deliberative governance and sortition-based governance.

References

• Decision Lego: A Primitive on Governance for the 21st Century

• Cosmos Citizens' Assembly Pilot Report: Cosmos Forum Report

• Optimism Collective Deliberative Processes Experiment: Optimism Governance Forum

Humanity stands at a critical juncture, confronting global challenges of unprecedented complexity that necessitate a collective approach to decision-making, spanning from local communities to the global arena. The existing governance processes, however, show signs of strain under these demands. Many of these processes, steeped in historical legacies, now grapple with the need for evolution, while emergent methods still falter in achieving effective implementation at a broader scale. Some methodologies even amalgamate the drawbacks of both traditional and contemporary systems. This underscores an urgent need to reinvigorate the discourse on governance, moving beyond the conventional paradigms of international relations and political theory.

This paper introduces a novel hypothesis: envisioning governance as a versatile “Lego game”, composed of “bricks” — each representing distinct governance processes and systems. These bricks, varying in shape and color, symbolize the diverse elements that can be assembled to construct comprehensive governance frameworks or on the contrary help in deconstructing existing governance structures to scrutinize and refine them.

Employing a multidisciplinary approach grounded in "fuzzy logic" (Cintula et al., 2021; Zadeh, 1975), this paper offers a unique perspective, blending theoretical insights with practical experience. The author's involvement in key governance fields — Aleatorian democracy since 1998, deliberative democracy since 2003, and the Web3/decentralized governance ecosystem since 2017 — provides a rich, self-reflective ethnographic backdrop for this exploration. Rather than claiming a completely novel methodology, this paper is structured as a synthesis of experiences, tapping into research and the practical application of innovative governance formats. It also aims to serve as a foundation for the proactive design of high-quality governance architectures, taking into account not only the processes themselves but also their contextual applicability.

This paper is the inaugural installment in a series that will delve deeper into the practical implementation of the 'governance Lego' in various real-life political settings.

Old solutions are not adapted to new problems

We cannot solve our problems with the same thinking we used when we created them. Albert Einstein

The tools we use for collective decision-making – also called governance – appear to have a major difficulty in coping with challenges of the 21st Century. The hypothesis I make in this paper is that the tools we use mainly date back to the 19th and 20th centuries and are not adapted to the 21st century’s society and its challenges. Not only do modern societies have profoundly evolved since then (buzzwords: Global population X10; digitalization; interconnectedness; Rising education level, etc.) but also Humanity is facing new challenges unknown to past societies, like Digitalization and Synthetic intelligence, Climate Change, growing inequalities at global level, loss of Biodiversity, polarization, technologies like genome editing able to change the very meaning of what it is to be Human, etc.

There is a clear lack of adequate institutions for decision-making and collaboration to effectively tackle those challenges. This situation translates into a growing gap between decision-making and the people those decisions are made for: citizens. An increasing part of the population feels excluded from the process or is objectively disenfranchised from it, which leads to even more blockages in the system. It generates a vicious circle of mistrust, harming the capacity of decision-makers, stakeholders, and populations to agree upon and then implement the needed solutions.

The focus of this paper is not to enter an in-depth analysis of why the tools are not relevant and/or effective. It is rather focused on proposing a framework for tools that could be up to the task. Therefore, I will approach the analysis of existing tools through a positive reformulation of what they should ensure to help solve our challenges. I will do this thanks to a set of criteria of quality. Before this, however, it is necessary to really understand what a decision-making process is.

What is governance? A typology

Who gets what, why, and how

During the classical democratic era in Athens, members of the Council of 500, or Boule – one of the main institutions of Athenian democracy – were chosen once a year from all citizens of Athens aged 30 or more by means of a random selection. Before they could perform their duties, the drawn “bouletai” had to pass a so-called “dokimasia” in front of the former Council. It was a test in which they were questioned about all sorts of topics (their family, values, etc.). Subsequently, they were confirmed or rejected by a hand vote. Once they were accepted, they had to serve for the year; their role was to prepare decisions to be put to discussion and voted in front of the Ekklesia, the citizens’ assembly (Hansen 1999).

In late 2020, the decentralized exchange Uniswap, which allows users to swap digital tokens, distributed its UNI token to its users (Kamarul 2020). This so-called “airdrop” was shortly after valued at more than €8.000,00. The method used to distribute the tokens was based on a snapshot of users (Ethereum addresses) that had interacted with the exchange. And each of these addresses received 400 digital “tokens”. To claim the tokens, users had to sign a transaction with their digital wallet. The tokens could then be traded or used in the governance system of the protocol to vote on decisions concerning the project. The vote is based on a weighted system (more tokens, more voices).

At first sight, these two examples couldn’t be more different from another. At second sight, they both illustrate a core feature of any Human Society: The distribution of resources through a process of governance. Lasswell (1935) called this politics and summarized it as: “who gets what, when, and how.” In this paper I will substitute an element of the list to make it: “who gets what, why, and how”. This change underscores the importance of understanding not just the distribution of resources, but also the underlying motivations and justifications that drive these decisions as we will see later.

What: Lots, functions, and decisions

The resources, the “what” can be called items and can be divided into three categories, following Elster (1987, 108):

Lots: These are tangible or intangible assets distributed through governance. Examples include material assets like a computer, or immaterial assets like a ticket to an event, or digital assets like the UNI token previously mentioned. Functions: Functions encompass roles or responsibilities that can be either non-political (like specific jobs or tasks) or political. Political functions typically fall into six categories: Legislative: Concern the rule-making processes. A perfect example is a parliamentarian. Executive: Focuses on the enforcement of these rules. Typically civil servants. Judiciary: Ensures the conformity of rules with overarching laws or constitutions and enforces sanctions for rule violations. A judge would incarnate this kind of function. Consultative/Advisory: Provides advice or suggestions. This could be the role of citizens in Citizens’ Assemblies for example. Elective: Corresponds to the function of choosing other function holders. For example, the board of a company can choose who is going to be the new Chief Executive officer. Decisions that are to be understood as the selection of a course of action (turning right or left for example). This category is too vast to create a clear typology. That being said, if we focus on core decisions on governance (decisions on how to decide), we can pin down the following major categories: Organizational decisions: creation and dissolution, membership (inclusion / exclusion, rights, and duties), rulemaking (constitutional process for metarule and legislative for normal rules), diplomatic (relations to other organizations), budgetary decisions: decide on incomes and expenses, functional decisions (more granular kind of decision within the limits of the organizational and budgetary boundaries. We can call them projects or programs).

Using this distinction, the selection for the Boule can be described as a consultative function as the Boule prepared the laws for the Ekklesia (see Blackwell 2003, Hansen 1999 or Headlam-Morley 1891). The Ekklesia had the legislative function. The UNI airdrop can be described as the distribution of a Lot, giving the right to participate in the legislative function of the UNI project. Worthwhile noting and maybe less than a coincidence, both required a token: In the first case it was the name of the citizen put on a piece of bronze and used in the Kleroteria (random selection machine) in the second case a digital ERC20 Token, a standard for digital tokens on the Ethereum Chain.

When and how: Selection procedures

A selection procedure can be described as a process by which a set of items is reduced to one: “moving from many options to just one” (Ullmann-Margalit 1977; Elster 1987). In the Boule example, the set was the totality of the male Athenian citizens over thirty. The reduction was performed by the combination of a random selection, an examination, and a vote. The procedure was repeated for each of the 500 Bouleutai. In the case of UNI the set was all Ethereum addresses and the reduction was performed on the basis of a single criteria: The interaction with the protocol.

There have been many attempts at creating a typology of procedures and it's still a field in discussion[1]. I have in the past created a typology differentiating the following 4 types which I will use in this paper: Vote or Psephos, market or Agora, random selection or Kleros and examination or Dokimasia (Vergne 2013).

How to distinguish these four procedures from another:

• The vote uses the aggregation of individual opinions as a means of reducing the set of options. In a vote, preferences are translated into suffrages that are added with the help of an “algorithm” (simple majority, proportional representation, absolute or relative majority, consensus, quadratic voting, etc.). This results in a decision.

• In a market procedure, individual preferences are matched (reduced) using a price which is the balance between preferences (offer and demand).

• In an examination, the reduction is done by checking if a list of prefixed criteria are met. The most known of these criteria are rotation, birth (or nationality, inheritance), needs and merits, or time (a queue for example). Since the relevant criteria are defined by the actors themselves, and are combined, it is impossible to conduct a systematic enumeration here.

• Random selection is different from the three previously described methods in that the transformation of the individual preferences into a collective decision bypasses the will of the actors: there is a moment of “uncertainty”. Random selection is a chance device (Alford 1958, 2) and reflects the “intentional choice to make the decision by a non-intentional mechanism” (Elster 1987, 108). The reduction is therefore based on the use of an aleatory moment.

[1] The exact number of types of procedures is still being discussed. Stone (2009, 232) distinguishes seven methods: “Lots”, “Vote”, “co-optation”, “auction”, “strict rotation”, “temporality” and “monopolized allocation” (which he divides into different groups: distribution on the basis of certain “merit”, “suitability” or “needs”). Elster (1987, 128 sq) distinguishes six types of procedures: “random selection”, “equal physical division”, “allocation” (three types: allocation based on “needs”, “productivity” or “merit”), “market”, “queue” and “status”. Kornhauser and Sager (1988, 483) suggest a division in four types: “vote”, “allocation” (on the ground of “merit” or “need”), “market” and “random selection”. Saunders (2008, 361) distinguishes between three methods: “selection”, “random selection” and “auction”.

Why: The narrative of decision-making

When Athenians selected their lawmakers by lot, they did not consider that they were chosen by the gods (Manin 1997). On the contrary, when they went to the Aruspice to interpret their future by reading the guts of a dead animal, they did believe that gods were involved (Bromberger & al. 1987).

This distinction is important as any decision-making has a “material” and “intellectual” dimension. The Intellectual dimension is both the way people justify the procedure and the way they interpret it:

Justifications: When making a decision people refer to “reasons why” they do it. For example, when selecting through an election they refer to transparency or competence. When recurring to random selection they put in front reasons like equality of chance or protection against corruption. The market is often justified as being efficient or scalable or able to exist without a central coordination mechanism. An examination is justified by reasons like need or rationality. We will see later that justification is the key to understanding governance systems.

Interpretation: There seem to be 3 main regimes of interpretation: The first one is finalist meaning that the output is seen as the consequence of the will of a superior entity (god, the gods, destiny, etc.) the second one is deterministic meaning that the output is seen as the result of a causal chain that can be traced. The last one is probabilistic meaning that the output is seen as being based on a statistical probability of the event occurring given the input conditions.

This dimension of decision-making may seem very logical and uncontroversial, but it has 1 major consequence: Decision-making is part of a narrative

None of the 4 decision-making procedures presented before have a “natural” or “intrinsic” property. If you claim to want to use a market-based decision-making because it is “transparent” but you let insider knowledge bias the mechanism, you are creating a non-transparent mechanism. If you choose your rulers based on their merit of being born in a cast or family (kings and queens) you are recurring to random selection, as birth is random. But you may well interpret it as the will of a superior force, in which case this process is for you based on merit.

Having this in mind allows you to open a large spectrum of design possibilities as you can play with the narrative and the decision-making tool proactively.

The procedural block as unit of account

When looking at real-life governance, this typology reaches its limits as no real-world process is a clean-cut “vote” or “market”. Any governance is embedded in a more general process. One way of approaching this is to consider “procedural blocks” as a series of procedural steps resulting in a decision. Taking the example of the Boule the block would start with exanimation (being a man over 30), continue with a random selection, and then finish with an examination (in front of outgoing bouleutes). This approach will allow us to imagine decision legos in a much more realistic way.

This part of the paper has shown that decision-making processes have no intrinsic characteristics. It doesn’t mean that they are all equals. There may be “bad” processes and “good” processes. As these terms are highly normative, I prefer to approach this question through the lens of the quality of the procedural blocks.

Governance should be inclusive, deliberative, relevant, and fair

Quality refers to the fact that something is judged satisfactory by the people interacting with it, regarding a series of criteria. In a hotel, quality could refer to the cleanliness of the room, the responsiveness of the personnel, the noise, etc. In this part of the paper, I focus on the criteria for a high-quality governance process. It is based on previous publications and inspired by many existing frameworks (Vergne 2011; Vergne 2014). As stated before, the unit of account is the procedural block.

Inclusion

A high-quality process should be capable of engaging everyone in the process as inclusion is the key to legitimacy but also to a higher cognitive quality of the decisions taken (Landemore 2014; Surowiecki 2004). So, inclusion is meaningful in terms of input and output. Inclusion has three dimensions that can be complementary or additional.

Quantitative inclusion

This means that as many people as possible are included in the process. Such an inclusion ensures a direct ownership of the process and its results. A good example of a potentially very quantitatively inclusive process are general elections coupled with universal suffrage as any citizen may participate.

Qualitative inclusion

This kind of inclusion means that the most diverse set of opinions and viewpoints are included in the process. This can for example be achieved by sortition (random selection) of participants or by carefully mapping and including representatives of very different positions.

Incentives

Incentives are a set of leverages that motivate participation. The higher the incentives, the stronger the probability of having both a large quantitative and qualitative inclusion. Incentives are commonly divided into 4 groups: Social, financial/material, political, and cognitive (Mazeaud and Talpin 2010).

Social incentives are top-down (group pressure and norms that “force” participation) or intrinsic (wish to be part of a Human group, sociability, joy). Social incentives relate to the very nature of Human beings as social beings (Bregman 2020).

Financial incentives are straightforward as they motivate participation against payment. This form of incentive is the backbone of industrial societies. It’s also a key feature of deliberative processes as participants are often rewarded for their participation based on the old tradition of the Greek obola. Financial incentives are of course the central feature of blockchain ecosystems. As Buterin (2018) puts it, the goal is: “to reduce social trust assumptions by creating systems where we introduce explicit economic incentives for good behavior and economic penalties for bad behavior.” Material incentives are another form of direct incentives. For example, the invitation to gather at a single place face-to-face all expenses paid is a very strong incentive that largely enhances inclusion.

Political incentives connect to the wish of Human beings to be part of the collective decision-making process, to influence it and/or to control it. They are the backbone of any decision-making process.

Cognitive incentives are based on the wish to learn and experience. It’s a very powerful motivation in any collective endeavor even if it is often ignored, downplayed or even mocked in many societal contexts.

Deliberation

A high-quality process can be measured against the argumentative process it goes through. Following Habermas, an ideal situation could be described as one in which the final consensus or state of a discussion is based on the force of the best arguments and not on power situations. A deliberative process relies on the following steps that are partially sequential and partly parallel or in loops:

Information: In this phase, the goal is to gather information from participants or from outside. It can be opinions, facts, or expertise.

Debate: The second core component of good deliberation consists of an exchange of diverse and/or opposed views, facts, and opinions on the subject matter. The goal of a debate is to confront different perspectives. It’s the key element of the judicial process. Having a debate is central to “building dissensus” or at least exploring it. It allows us to open the discussion and make sure that no key elements are overlooked. The debate can arise naturally or can be provoked through a facilitated process to overcome social complacency.

Dialogue: In this phase of deliberation the goal is to go from collected intelligence to collective intelligence in the sense that the information and the debate are melted into an exchange of arguments and the process of finding common ground. Common ground can be an agreement or an agreement on a disagreement.

Aggregation: The focus of this phase is to end the process. Aggregation can be made through one of the four decision mechanisms evoked above. It may be a consensus, a majority vote, a market process, an aleatorian process, or a mix (onchain vote with tokens which mixes market and vote for example).

Results: Once the cycle is over, the deliberation produces an outcome which is a result. This can be the final result or a result opening a new cycle of deliberation.

All steps can be iterative: A debate often gives new information, a dialogue opens debates, aggregation may result in the need to open a new debate, etc.

It is worth noting that this criterion may seem irrelevant to market mechanisms. This view ignores the fact that deliberation is taking place “around” the selection procedure, as part of a procedural block. The agreement on a price is the result of the deliberation of parties on their perceived quality of the product, or good or else.

Relevance

A highly qualitative process should be relevant. It means that it should be in line with what participants want to achieve. This ensures that the process fits the purpose and is neither over-engineered nor under-engineered. It also makes sure that resources needed are evaluated, so the decision-making is efficient. There is no need to have the full participation of everyone to decide the color of a bank or the background of a website (you know those meetings with the front-end team and the management).

A relevant process is also enforceable: It has an impact on the community and has consequences.

Fairness

Finally, a highly qualitative process demands fairness. The first subcriteria of fairness is reflexivity meaning that participants should be able to talk about the process itself and change it if needed (Dryzek et al. 2017).

Transparency is the second key characteristic of a fair process as it levels out power asymmetries.

The final sub-criteria of fairness considers the sustainability of the decision-making process understood as societal, ecological, and economic sustainability. Sustainability is also connected to its temporal dimension (how does it impacts the 7th generation) and its legacy (what will stay).

In this section, we have gained clarity on the quality of a procedural block of decision. We have gone from a decision-making mechanism to a decision-making procedure, to a decision-making procedural block.

This architecture is very useful, but it misses a broader perspective because decision processes do not exist in a theoretical bubble, they are part of a governance system or ecosystem. Therefore, I now turn to a new typology.

A fuzzy but relevant typology of governance systems aka polities

The classical Aristotelian typology of governments differentiates between pure and mixed systems and between two dimensions: Who governs and for the benefit of whom (Hansen 1999). This is a highly effective way of classifying governance systems which can also be tagged as polities following the old definition of a polity being a political system. I will use the typology in a fuzzy inductive-deductive empirical-theoretical typology of ideal types of governance systems. By this, I mean a typology that mixes some real-world use cases with more theoretical ones, and “seems right” because they are ideal types, and we can understand what they mean.

https://opensea.io/assets/ethereum/0x006E88efc6f0EE43f112d66b1d1331800e8a7cb5/1

To illustrate the typology let’s take a couple of examples. First, if we look at one person governing it its own interest. The despot as ideal type seems quite relevant. If we now turn to some people governing for the benefit of everyone – or at least the immense majority – based on their birth. This rings like a good old aristocratic system. If that same limited group of people governs for the benefit of a majority based on merit or competence, we have a Technocracy. Replace merit by a vote (the few being elected and re-elected) and we get the ideal type of representative democracy. A government in which some randomly elected citizens govern to the benefit of all is an aleatorian democracy.

Existing governance systems are of poor quality

This paper is not the place to dive into an exhaustive panorama and assessment of all governance models presented above as it aims at setting the primitive. A series of following case studies will dive into specific polities and/or procedural blocks, their limitations in terms of governance, and proposals to improve them. This section focuses on shedding light on the most current/existing models with the perspective of the criteria of quality. And without any delay, it can be said that they perform poorly.

If we start with the most spread models, we find many nation-states (around 90) governed as some kind of autocracy, plutocracy, or nepotism. Around 90 countries of the world live under a form of hybrid regime or flawed democracy or technocracy, with a reduced group of people governing in the name of all (although not really for the benefit of all) (Democracy Index 2022). According to the Democracy Index, only 21 countries qualify as full democracies. Even there, the extent to which government is “for all” is questioned. If we look at governance models outside nation-states we find mostly corporations or hierarchic organizations like churches in which the majority of the people serve a reduced group governing in its own interest (shareholders).

These forms are of poor quality because they are not inclusive, they are not deliberative, and surely not fair. They are in many cases relevant because they can be efficient and can scale. Nation-states have shown how well they managed the industrial revolution. However, they are not sustainable as they are not in a capacity to integrate long-term thinking and are based on an extractive model and not a regenerative one.

Luckily, many people have been working on improving this situation and finding, testing, and improving new bricks of decision-making, new procedural blocks, that would live up to a high quality of decision-making: Nice, colorful, decision legos.

Decision legos

We have delved into the study of decision-making, its components, some possible criteria of quality, and the reason why many existing constructions are of poor quality. But there is hope, as deconstructing a Lego into its pieces is the best starting point to to build new models. And with those bricks come millions of combinations, billions of possibilities to play with decision legos to create inclusive, fair, relevant, and deliberative decision-making processes. It is worth noting that this approach has the immense advantage of being fractal: You can scale the Lego bricks up or down at any level and they look the same. A decision within a team of three or a crowd of millions can be dealt with through a comparable process.

This paper aims at proposing a framework, a primitive. It is not the place to delve into the details of designing concrete models of governance or passing them in review. Also, many questions on piloting, and implementing new sets of legos remain unaddressed. It may be that some bricks can not be plugged on some others. That too many colors make it difficult to read the ensemble.

In an upcoming and ongoing series of use cases, I will focus on precise ecosystems, polities, or procedural blocks. Particularly, I will start exploring such ecosystems and combinations that seem to be the most promising to achieve better decision-making in the 21st century quickly and that have been piloted and developed with an expressive goal to counteract the shortcomings of the most widely spread polities. These are on one side bricks leveraging deliberation and sortition and on the other side bricks activating and mobilizing Web3 principles and infrastructure.

Humanity is mature to engage in new ways of making collective decisions that are aligned with planetary boundaries and respect our very nature as social animals. Engaging in the reflection, prototyping, and testing of decision-making legos seems to be a productive path to let a new renaissance flourish.

Web3 presents a unique opportunity to redefine value distribution and community engagement through mechanisms like airdrops. However, the challenge lies in crafting these mechanisms to be fair and effective. This note proposes a solution rooted in democratic deliberation: utilizing citizens' assemblies to design the criteria for airdrops, ensuring they reflect the collective will and diversity of the community they aim to serve and reducing the attack surface for critics.

The Need for Inclusive Criteria in Airdrop Design

Airdrops have the potential to democratize access to digital assets, foster participation, and distribute governance across a decentralized ecosystem. However, without meticulously designed criteria, airdrops risk reinforcing existing inequalities, overlooking diverse community segments, or failing to engage the community meaningfully. In the worst case and as we have seen again and again, the airdrop turns the community against the project itself. The question then arises: how can we design a process to define airdrop criteria which is fair for all stakeholders?

Citizens' Assemblies: A Mechanism for Collective Wisdom

Citizens' Assemblies offer a blueprint for engaging a microcosm of a community in deep, deliberative decision-making. By randomly selecting members to reflect the broader community's demographics, these assemblies can gather diverse perspectives, mitigate biases, and reach consensus on complex issues—like the criteria for airdrop eligibility and distribution. Assemblies have been around for 50 years but have exponentially taken off in the past 5 years: They are used by many public authorities and private actors to improve decision-making in complex environments. The European Commission has based its full Web4.0 strategy on such an Assembly, the French government has recently shaped its climate and end-of-life policies thanks to Citizens’ Assemblies. In the US state of Oregon, CAs are coupled with referenda. Meta has already run two rounds of CAs (called community forums) to craft internal policy (for example on AI). In Web3, we have recently deployed the first pilot in the Cosmos Ecosystem around the condition to enter and leave the Atom Economic Zone. CAs are powerful because they combine cognitive diversity (through random selection of participants), procedural strength (through a process of informed and facilitated discussion), and legitimacy (through the non-partisanship, efficiency, and transparency of the process).

Leveraging Citizens' Assemblies for Airdrop Criteria Design

A CA is a perfect fit for airdrop criteria design for the following reasons:

1. The random selection of participants allows to balance out the overweight of the loudest crowd and make sure to include the silent majority.

2. The process of learning and expert hearings allows to bring in the expertise and experience of past airdrops teams to reflect on their successes and limitations.

3. The process is open: This cuts short to speculations, exaggerated farming, FUD (e-beggars), and FOMO (which almost always brings frustration).

4. At the same time, it is better than point systems that mainly reward bigger players and have a plutocratic bias. Point systems also often ignore past usage.

5. It allows the distribution team to get a very precise image of their community when their community thinks: CAs are “what the people think when they think”. They go out of their bubble.

6. The random selection + process is a very powerful defense against FUD from professional farmers as the legitimacy of random selection + the open process cut the accusations of backdoor deals, untransparent criteria crafting, etc.

7. Citizens' assemblies can deliberate iteratively on what constitutes fair eligibility criteria for airdrops, considering factors such as prior engagement, contribution to the ecosystem, and even measures to ensure newcomers have opportunities to participate. This process leads to the development of innovative criteria that balance rewarding past contributions with fostering future engagement and diversity.

8. Ensuring Inclusivity: One of the most critical roles of citizens' assemblies is to ensure that the airdrop criteria do not inadvertently exclude underrepresented groups. Through deliberation, assemblies can identify and mitigate potential biases, ensuring that the airdrop serves as a tool for enhancing diversity and inclusivity within the community.

Conclusion

I propose piloting a Citizens’ Assembly process to support badgeholders in the assessment and voting process of the 4th round of Retro public good funding. This assembly would allow us to overcome the most blatant shortcomings of the RPGF3 process. It would refocus the process from collected intelligence to collective intelligence.

What is the RPGF of the Optimism collective

For readers not familiar with the Retro public good funding (RPGF) process, here is a short introduction. For others, go directly to the next paragraph.

The Optimism Collective associated with the Optimism blockchain, is a community dedicated to supporting public goods through innovative funding strategies. The idea of retroactive Public Goods Funding is to reward projects or individuals after they have created something that benefits the public, especially within the blockchain or Ethereum ecosystem. The idea is to provide incentives for work that has already proven to be valuable, as opposed to traditional funding mechanisms that invest in potential future outcomes. The RPGF focuses on Public Goods meaning projects that are non-excludable and non-rivalrous. In the context of the blockchain and Ethereum, this can include open-source software, educational resources, community-building initiatives, and other projects that contribute to the ecosystem's health and growth.

The decision on which projects or individuals to fund is made through a decentralized process, involving so-called badegholders who are a selected group of community members acknowledged for their reputation in the community.

Perceived limitations of RPGF 3 process

The third round of Optimism's Retro Public Goods Funding (RPGF) is now closed and is facing several criticisms and challenges:

One key point of contention is the eligibility of projects with substantial financial backing, including those funded by venture capital. Critics argue that such projects, already financially supported, should not be eligible for RPGF funds, which are intended for contributors without such backing.

Feedback from the Optimism Collective community also highlights issues with the RPGF process. For instance, there was confusion over the distinction between individual and team grants, with concerns about potential "double-dipping" when both received funding. The use of lists in the voting process was another contentious topic. While lists made voting easier for badgeholders, they may have encouraged lazy voting and conformity, with people copying lists without a thorough review of the projects.

Furthermore, there is a call for more accountability among badgeholders, with suggestions that those who vote irresponsibly or without proper rationale should face consequences. The feedback also emphasized the need for a more transparent and rigorous project application process to prevent spam and ensure high-quality submissions.

Deliberative governance to the rescue

Implementing a citizen's assembly of badgeholders could be an effective solution to overcome the limitations observed in the third round of the Optimism Retro Public Goods Funding (RPGF).

1. Diversified Decision-Making: A citizen's assembly would bring together a diverse group of badgeholders with different perspectives and expertise. This diversity can help mitigate the risks of conformity and groupthink that arise from reliance on lists and popular opinions. It encourages more holistic and varied viewpoints in the decision-making process, leading to a more representative and balanced selection of projects.

2. Enhanced Accountability and Transparency: Such an assembly would operate under a framework that ensures accountability and transparency in its decision-making processes. Badgeholders would be able to collectively articulate their rationales for voting, fostering a culture of responsibility and thoughtful consideration. This setup also allows for the community to better understand and trust the decision-making process.

3. Structured Deliberation: A citizen's assembly model typically involves structured deliberation, where participants are provided with detailed information about each project and have the opportunity to discuss and debate their merits and drawbacks. This structured approach ensures that each project is given fair consideration and that decisions are based on a thorough evaluation rather than superficial factors like brand recognition or popularity.

4. Reduced Workload Per Individual: By distributing the workload among a larger group, the individual burden of assessing and voting on a high volume of projects is lessened. This can lead to more effective and less rushed evaluations, as each member of the assembly can dedicate appropriate time and attention to a smaller number of projects.

5. Potential for Innovative Solutions: A citizen's assembly of badgeholders, with its diverse composition and structured approach to decision-making, could also be a breeding ground for innovative solutions to emerging challenges within the RPGF process.

This approach aligns with the principles of decentralized governance and community involvement that are central to the ethos of the Optimism collective. It offers a way to refine and improve the RPGF process, making it more equitable, transparent, and effective in supporting public goods within the Optimism ecosystem.

The assembly would demand 3 days of work. It could be centralized and take place in one single place or be distributed in a couple of hubs depending on the place of living of the badgeholders.

2024 marks the 50th anniversary of the first known “minipublics”. Since then, the discussion about the “right” process for selecting citizens through random selection for a political office (aka sortition) is ongoing. Major progress has been made in the theoretical and practical field. We have seen a strong community of practitioners gather; an incredible amount of research being made. In some cases, we have even seen a full circle: For example, in Germany door to door recruitment was used in the very first Planning cells and is again gaining traction as a preferred method after four decades of ignorance.

Despite those progresses a core question remains: What is the best method to gather a representative sample of the citizenry to participate in a deliberative forum?

My hypothesis is following: There is none.

Instead, we need to embrace an intersubjective approach of sortition. An approach that puts at its core the realization that we do not have the technology nor the data to achieve an objectively clean sortition. More than that: Maybe we shouldn’t even try.

The argument of this essay goes as follows: What is sortition really; Why it is a socio-political process; Why we should embrace this and be modest about it; How we can work out a process to get a representative group people following an intersubjective approach.

What sortition is

To take decisions, Humans use a so-called selection procedure which can be defined as a process to go from many options to one, using a mechanism. There are four main categories of selection procedures: voting, market, examination (criteria), and random selection. How to distinguish these four procedures from another:

1. The vote uses the aggregation of individual opinions as a means of reducing the set of options. In a vote, preferences are translated into suffrages that are added with the help of an algorithm (simple majority, proportional representation, absolute or relative majority, consensus, etc.). This results in a reduced set of options.

2. In a market procedure, individual preferences are matched (reduced) using a price which is the balance between preferences (offer and demand). When offer and demand agree on a price, the reduction is done.

3. In an examination, the reduction is done by checking if a list of prefixed criteria is met. The most known of these criteria are rotation, birth (or nationality, inheritance), needs, merit, or time (a queue for example). Since the relevant criteria are defined by the actors themselves, and are combined, it is impossible to conduct a systematic enumeration here.

4. Random selection is different from the three previously described methods in that the reduction of the set of options bypasses the will of the actors: there is a moment of “uncertainty”. Random selection is a chance device (Alford 1958, 2) and reflects the “intentional choice to make the decision by a non-intentional mechanism” (Elster 1987, 108). The reduction is therefore based on the use of an aleatory moment.

Sortition is the use of random selection to select members of a decision-making body: Juries, Assemblies, Councils, etc.

Random selection can of course be used to reduce the set of options in other settings: In a lottery it usually determines who get a prize. In a coin flip it decides upon two options, for example who is going to kick the ball first in a football game.

Why is sortition a social and political process

Selection procedure, hence random selection and sortition are not artefacts isolated in space and time. They are embedded in a societal context. Each selection procedure not only is a mechanism but also an intellectual operation. When taking a decision people refer to “reasons why” they do it. For example, when selecting through an election they refer to the transparency, or the competence. When recurring to random selection they put in front reasons like equality of chance or protection against corruption. Market is often justified as being efficient or scalable or able to exist without a central coordination mechanism. Examination is justified by reasons like need or rationality. Random selection is often justified to be a way of allowing diversity, and to avoid vested interests or corruption. It means that any instance of sortition is based on the wish to achieve a particular goal.

Further, any selection mechanism is enshrined in a mental model of interpretation of which there seems to be 3: The first one is finalist meaning that the output is seen as the consequence of the will of a superior entity (god, the gods, destiny, etc.) the second one is deterministic meaning that the output is seen as the result of a causal chain that can be traced. The last one is probabilistic meaning that the output is seen as being based on a statistical probability of the event occurring given the input conditions. The third regime is the most shared nowadays.

“The names of all guild members over thirty years old were put in eight leather bags called borse. Every two months, these bags were taken from the church of Santa Croce, where they were ordinarily kept, and in a short ceremony drawn out at random. Only men who were not in debt, had not served a recent term, and had no relation to the names of men already drawn, would be considered eligible for office.”

Signoria of Florence, Wikipedia

In conclusion, we can say that sortition is the product of a design both in the tools that are used as well as in the justification that are given to its use. We can therefore embrace with confidence an intersubjective approach of sortition.

Why are current sortition processes unsatisfactory

Current processes of sortition are not satisfactory because they pretend to ignore the deeply relative nature of any selection process. Instead, they try to objectivize selection as if independent, objective criteria would exist to reach representativity. Let’s take 3 examples:

1. Age. In this case the goal is to mirror the age structure of the general population. But this omits the fact that young people are immediately rejected for being too young to participate (until 16, or 18 or any other age decided as being the right age). And the oldest people will simply not be able to participate because in high proportion they will be physically or mentally impaired (think 95+). Also, respecting the age pyramid may bring damaging biaises in countries with major imbalances like Western countries (a lot of old people) or countries like Morrocco with a very high number of young people (around 25%). This may be representative in the demographic dimension but completely unfair in terms of social justice or opinion biases in the group.

2. Gender: The classical dichotomy man/women is being questioned in many ways. More and more Human beings identify themselves on a more fluid scale of gender identities. How to treat this in a sortition process? Add one category (other)? But isn’t this a discrimination? So should we add 5 or 6 other options? And how to capture the % of the population identifying with those categories? And what if in a specific socio-cultural context this very option of going beyond two gender is seen as a bias by the majority of the population and leads potential participants to refuse to participate?

3. Attitudinal criteria: Many processes include some attitudinal dimension like a position on the topic at stake. For example, during the recruitment for the European Citizens’ panels, potential participants where asked about their view of the European Union from skepticism to support. This is a very interesting criteria as it allows to capture a diversity of views and can be related to a set of statistically representative data (Eurobarometer). But again: who is responding to Eurobarometer and can we be sure that there is not bias here. We have seen in the 2016 US American elections (and in many other contexts) that “representative” opinion polls can be very wrong in the way they reflect the actual views of the population. It means that the “universe” against which representativity can be measured is questionable. Moreover, attitudinal criteria are a rabbit hole: If we start adding one, why not add a second and a third and hundred more to better reflect the population?

So: Do we need to give up and flip back to election through vote as the best proxy for selecting a representative decision-making body?

The answer is no.

Intersubjective sortition

Whenever we want to use sortition, we will have to agree on the process that will make it representative in the eyes of the involved parties. This is what I call intersubjective. At each step of reduction, choices are being made. These choices are political, cultural, legal. They are different for each process and subjective. It means that there is a flaw in any sortition process because it relies on a process of design. That is true for any selection mechanism: deciding that women or children vote or do not vote is subjective.

“I am here as a soldier who has temporarily left the field of battle in order to explain - it seems strange it should have to be explained - what civil war is like when civil war is waged by women. I am not only here as a soldier temporarily absent from the field at battle; I am here - and that, I think, is the strangest part of my coming - I am here as a person who, according to the law courts of my country, it has been decided, is of no value to the community at all; and I am adjudged because of my life to be a dangerous person, under sentence of penal servitude in a convict prison”.

Emmeline Pankhurst. Freedom or death, speech delivered in Hartford, Connecticut on November 13 1913

Unlike the history of voting rights though, which has been a rather linear process of including more and more people, the discussion on sortition is more convoluted. Objectivity is even more relative. This is because of the justifications chosen to use random as a selection mechanism. Some arguments:

1. Equal chance: Such a justification would mean that we need to use pure random on the universe itself (everyone) to abide by the theorem of large numbers. This works only if the data on the population is complete from the outset and the number of participants high enough. In practice: impossible. And desirable? Do we want to create a database of all Human beings so complete that it allows to run that sortition? Maybe, maybe not.

2. Equal opportunity: Such a justification means to have a proactive process to overrepresenting or excluding some sectors of the population. But who is to decide? In the Panels of the Conference on the Future of Europe, it was decided that young people should represent 33% of the participants. This number is not statistically representative of the population as the 16-25 represent around 12% of the population and the 0-25 represent around 25%. The 33% came out of a negotiation process.

3. Inclusion: Such a justification means that we skew the universe and create a population by adding exclusion/inclusion criteria before applying any sortition. It can be age, but it could be sociodemographic criteria too: poorer people being overrepresented to counterbalance the power of rich people in society. Here again: The basis for the sortition is a political process.

As a conclusion, the best way to address this elephant in the room is to proactively propose an intersubjective approach to sortition: Sortition is a process of selection based on an intersubjective design: Stakeholders need to agree on the right set of justifications, criteria, and steps of the process for it to be legitimate. In this view, there is no intrinsic quality of the random selection but a procedural one.

Thinking in more practical terms, this process could be organized in six steps:

1. Go from the universe to the population: In this step, the goal is to define the possible options amongst the infinite set of options. For example, we can decide to focus only on living Human beings and exclude future generations and plants and animals. We can be more restrictive from the beginning and decide that only Human beings living in a specific country should be considered. Taking again the example above, we may decide to consider two genders, or 4, or more. At this stage, it doesn’t matter to know the proportion of each group in the population. So: We don’t need to know how many Human beings we have living in the country. We only need to decide if this category is part of the considered population. As an extreme example: To work on the topic of prostate cancer prevention policy, we may want to reduce the population for the sortition to male over 40.

2. Once the population is defined, we need to characterize it with the help of data: This is where we need to agree on a legitimate set of data as being good enough to represent the diversity of the population. In some context and for some data this may be easy. For example, the countries of the EU have good dataset on age. But, I once met an old man in Chile who was born in 1801 (so he told me at least). So, this may become tricky and the most important is to have agreement among stakeholders that the base data is legitimate.

3. Once the dataset is agreed upon (characterized population), it’s time to agree on the sample, meaning the subset of the population which is going to be the target for the group of participants. If my dataset says that I have 25% of my population aged between 15-25 and I have a group size of 100, my sample will have to contain 25 Human beings aged 15-25. At this stage, we may decide to skew the sample to reach certain goals. For example, we could agree to overrepresent the 15-25 age group to also consider the 0-14. But this will mean that we underrepresent other categories as we can’t go over 100%. The sample should be a number equal to the number of actual participants for each category (target).

4. The next step is to deploy the campaign of recruitment and reach out to real Human beings and profile them according to the criteria. This helps create the pool of candidates. Here again there will be a skew: Some target groups will be less responsive to the campaign: They will not open the door, will not answer the phone, will send their parents, sisters, sons. The random selection will be distorted, and it will be very difficult to know in which proportion. It is important to agree on the rules to handle such deviations (do we accept a person from the same household that allows to fulfill a slot in the pool?). It is also the major step in which the channels for the campaign need to be agreed upon: Random phone number generation? Letters? Door knocking? Social media? A mix? If a mix: What proportion of the pool should be recruited by each channel? The size of the pool should also be agreed upon: Do we want twice as much as candidates as participants to have a 1:1 reserve? Are we happy with 1,5 the size (meaning that we can accept 50% no show).

5. The next step is to select the actual participants from within the pool. This can be done by matching the targets of the sample with the candidates in the pool. This step is best done with software tools like panelot, StratifySelect, random.org, or equivalents. It can help create a main list and a reserve list.

6. As a last step there is a management process by which no show, and blatant deviations from the targets can be corrected as the process unrolls.

Conclusion

Sortition should be considered as a social and political tool and not as an objective artefact with intrinsic properties. Doing this, unlocks a design space for better concrete sortition processes, making them more transparent, legitimate and resilient.