AI Disclosures Project | Fair, Participatory AI Markets (original) (raw)
Open, Participatory, Composable AI Markets
01 / Mission
Turning the notion of inspectable, participatory, value-creating AI markets into functioning reality. We bring a political-economy lens to the technical architectures and mechanisms that shape who benefits from AI and how.
AI systems are built on the creativity and data of humans and the open web, yet nothing attributes or rewards that value back to its sources. No such set of protocols exists — and without it, new human+AI markets cannot sustainably emerge. We need new market institutions, technical standards, and economic mechanisms that ensure value circulates broadly rather than becoming trapped inside a handful of gatekeepers.
02 / How We Work
Protocols & Architecture
Modular, interoperable architectures built on open protocols, such as MCP, can support commercial AI markets that distribute value broadly.
Mechanisms
Well-designed mechanisms can align competing incentives around shared growth, much as search and click-based advertising once did for the open web.
Prototyping
Working with builders, platforms, and policymakers to shape the standards and tools that get actually adopted. Making market design infrastructure enforceable, inspectable, participatory, and governable.
Convenings & Partnerships
Building shared technical standards and norms across communities. Bringing together economists, AI labs, platforms, builders, and policymakers who can design and implement change.
03 / The Big Picture Questions
Why Protocols? What does this have to do with disclosures?
At the AI Disclosures Project, we see disclosures through the lens of networking protocols and standards. Every networking protocol can also be thought of as a system of disclosures; these are far more than warning labels or mandated reports.
They are a form of structured communication that enables independent, decentralized action. The race for first-mover advantage by large centralized AI providers suggests a hub-and-spoke railroad design, while a world of open-weight AI models connected by new modes of standardized communication could look more like a road system, or today's World Wide Web.
If we want a world where everyone — not just AI model developers and those building on top of their centralized networks — is able to innovate and offer their work to others without paying a tax to access centralized networks, we need a system of disclosures that enables interoperability and discovery.
In this approach, protocols, as a type of disclosure, can architect healthier AI markets — not after things are already too far gone, but through operating as foundational "rules of the road" that enable interoperability.
"We need to stop thinking of disclosures as some kind of mandated transparency that acts as an inhibition to innovation. Instead, we should understand them as an enabler. The more control rests with systems whose ownership is limited, and whose behavior is self-interested and opaque, the more permission is required to innovate. The more we have built 'the rule of law' (i.e. standards) into our systems, the more distributed innovation can flourish."
— Tim O'Reilly
You can't regulate what you don't understand. And right now, critical information about how AI systems work, what data they use, and how they make decisions remains hidden inside corporate black boxes.
Guard against AI's enshittification. Cory Doctorow's term captures how platforms start out serving users, then shift to serving business customers, and finally optimize for extracting value from both. Without transparency about operating metrics, we won't know when AI systems begin this transition until it's too late.
Disclosures as a language of benchmarks. Just as accounting standards created a shared language for understanding business performance, we need disclosure frameworks that let us compare AI systems against meaningful benchmarks: not just capability metrics, but measures of fairness, safety, and alignment with user interests.
Disclosures shape market structure. The choice between open and closed disclosure regimes determines whether AI markets evolve like the open internet (where anyone can participate) or like railroads (controlled by a few gatekeepers). Disclosure standards, when designed well, become the protocols that enable competitive, innovative markets.
04 / Featured