From Broken Token Models to Market-Driven Governance: Interview with Umia CEO Francesco Mosterts
The gap between fast-moving AI agents and slow traditional capital formation is no longer sustainable. Founders can now build and scale agentic ventures at unprecedented speed, yet raising capital, structuring governance, and aligning incentives still drag projects down with delays, misaligned tokens, and outdated models.
Fresh off research into onchain capital formation and decision markets, Olayimika Oyebanji sat down with Umia co-founder Francesco Mosterts to discuss why today’s tokens are broken, how continuous market-driven governance can fix it, and what a better system looks like for AI-enabled teams. Let’s dive in.
Can you briefly tell us about yourself and your route to web3?
I came into crypto through markets. I was working in trading, thinking a lot about execution, pricing, and how markets coordinate capital and decisions. What stood out to me was that crypto had all the right primitives to build better systems, but they weren’t actually being used where it mattered most.
I started Chainbound as an Ethereum-focused R&D company to explore how these systems could actually be applied to strengthen the ecosystem. Through this work, it became clear that there was a huge gap in how capital is allocated and how decisions are made. If crypto is going to matter, it has to improve how organizations are funded and run, not just how assets are priced.
How would you describe the current gap between how fast AI-enabled teams can build onchain and how slowly traditional crypto fundraising and governance still operate?
There is apparently a fundamental mismatch. Teams can go from idea to product in a matter of days, with AI compressing the entire build cycle, design, engineering, and even iteration, into something that looks more like trading than traditional company building.
However, capital formation hasn’t caught up yet, and still involves slow raises, fixed rounds, opaque allocations, and governance that kicks in after the fact, yet still doesn’t really do anything. So you end up with the strange situation where building is continuous, but funding and decision-making are static. In my opinion, that gap is becoming the biggest industry bottleneck.
Why do you describe capital formation as a major bottleneck? What makes today’s token models fundamentally broken for scaling new projects?
The core issue is that tokens are disconnected from outcomes. They don’t give meaningful control over capital, enforce decisions, or create a tight feedback loop between what a project does and how participants are rewarded.
On top of that, the fundraising process itself is extractive. Capital is raised upfront, but tokenholders have very little influence beyond low-signal voting. With this, trust breaks down, and users hesitate to fund early-stage projects because they don’t know how capital will be used or whether decisions will actually be followed through. So capital formation slows down, even as the pace of building accelerates.
Founders can now launch ventures from code at a breakneck speed and get legal structure plus treasury in days rather than months. What are the biggest practical challenges you see teams still facing when adopting this model?
The infrastructure to launch exists, but the coordination layer is still missing. You can spin up a project, deploy contracts, and even set up a treasury quickly. But the moment real capital comes in, you run into the same problems: How is that capital allocated? Who decides? How do you enforce those decisions?
Most teams fall back to informal processes or traditional governance models that don’t scale. So the challenge becomes operating in a way that remains coherent as the project evolves.
With AI agents capable of autonomous management, how do you see governance and fundraising evolving when agents themselves can participate in or even run these processes?
Agents make the current model even more fragile. If you have autonomous systems deploying capital and building products, you can’t rely on slow, human coordination mechanisms like voting or offchain discussions. What you need are systems where decisions are continuously priced and resolved in real time.
Agents can then participate in those markets, allocate capital based on expected outcomes, and react instantly as new information comes in. In that world, governance is an always-on process rather than a periodic event.
Looking at the broader crypto and AI intersection, do you expect decision markets and continuous funding to remain a specialized tool, or will they become standard infrastructure for onchain organizations in the next 3–5 years?
I think they become standard because they’re the only thing that actually scales with how fast systems are moving. If building becomes continuous and autonomous, then capital formation and decision-making have to match that. Otherwise, you’re introducing friction at every step.
Decision markets and continuous funding are the natural endpoint of turning organizations into market-driven systems.
How do decision markets differ from traditional token voting in practice, and what problems in DAOs and governance do they aim to solve?
Token voting is expressive but not binding in a meaningful way and with no incentives for participation. Participation is low, outcomes are often ignored, and there’s no cost to having an opinion. That leads to weak signals and poor decision-making.
Decision markets, on the other hand, flip that. Instead of asking people what they think, you ask them to price outcomes. Capital is put behind decisions, and the market aggregates that information into a clear signal. More importantly, the outcome is enforced, and it directly determines how capital is used.
What are the key lessons you have learnt about capital formation and scaling businesses in the age of AI?
The main lesson is that speed changes everything. When iteration cycles compress, the bottleneck shifts away from building and towards coordination. The projects that win aren’t just the ones that can build fast, but the ones that can allocate capital and make decisions just as quickly.
The second lesson is that alignment has to be structural, not narrative. You can’t rely on good intentions or community sentiment. The system itself has to enforce alignment between builders and participants.
And finally, markets are still the best coordination mechanism we have. The question is where you apply them. We’re starting to see that they’re not just useful for trading assets, but for running entire organizations.
Any parting words?
Crypto spent the last decade making assets programmable. Moving forward, next phase need to be making organizations programmable. If we get capital formation and decision-making right, everything else compounds from there.
