Funding AGI Research and Development as an Open Resource for All of Humanity

Abstract

DAOAGI exists to fund the research and development of Artificial General Intelligence as an open resource available to the entire human society. Everything it finances will produce open-source software and open-weight models, including the research and the entire knowledge.

Open AGI work is chronically underfunded compared to proprietary efforts, and existing grant mechanisms are opaque, slow, and weak at verifying outcomes. DAOAGI is being formed as a decentralized autonomous organization to fund that work through transparent, community-governed mechanisms.

This founding whitepaper states the mission, frames the problem, acknowledges the open questions any honest attempt at this must confront, and sets the direction for the design work ahead. The detailed mechanisms — tokenomics, governance parameters, legal structure, verification layer, resilience architecture — will be defined iteratively with the ecosystem in subsequent versions. We are publishing early precisely because the hardest problems cannot be solved in isolation; they require the people who will build, fund, review, and depend on this system.

1. Foundation Statement & Mission

Our mission is to fund the research and development of Artificial General Intelligence as an open resource available to the entire human society. Everything financed by DAOAGI will produce open-source software and open-weight models, including the research and the entire knowledge.

We pursue this safely. The architectures, models, weights, methods, and knowledge that will shape humanity’s future must remain a public good — not locked inside corporate or state silos, and not reserved for the DAO’s contributors. DAOAGI is being built as a neutral institution intended to:

Fund AGI research and development based on verifiable progress, not promises. Every financed output is released as open-source software, open-weight models, and the complete research record.

Govern capital through transparent, stake-based voting that evolves with its contributors.

Capture a portion of the value it helps create and reinvest it, without restricting public access to software, weights, papers, data, or methods.

Operate with a clear legal structure and an irrevocable openness commitment that protects both researchers and the public.

This document is the founding statement. The system that turns the mission into a working protocol will be designed openly, with the community, in the versions that follow.

2. Problem Statement

Funding for open AGI is fragmented across foundations, corporate CSR budgets, and ad-hoc sponsorships. Evaluations are often opaque, and funds are disbursed upfront with little recourse if deliverables are not met. This discourages long-term investment and pushes top talent into closed labs that have no obligation to share software, weights, or knowledge. The result is a tragedy of the innovation commons: the most consequential technology of our era is being built behind closed doors while public alternatives starve.

Three structural failures compound the funding gap:

Opacity. Decisions are made behind closed doors; applicants have little visibility into why they were rejected or how to improve.

Disbursement without accountability. Funds flow upfront; when milestones slip, there is rarely a mechanism to claw back or redirect capital.

No sustainable feedback loop. Successful open-source outputs generate enormous commercial value for someone else, none of which flows back to fund the next round.

DAOAGI exists to fund AGI as an open resource for all of humanity. These three failures are why that work currently starves; the rest of this document, and the open design work it initiates, is how we attack them.

3. Acknowledged Weaknesses & Open Questions

Any honest whitepaper in this space must begin by admitting what it does not know. The problems below are not footnotes; they are the central design challenges we expect to work through with the ecosystem. We list them upfront so contributors know what they are joining.

3.1 Funding Scale vs. Frontier Labs

Open-source AGI is a billions-per-year problem. The mechanisms DAOAGI can credibly deploy at launch (donations, bonding-curve fees, IP-service fees) will not, on day one, match the capital base of frontier labs. Whether DAOAGI can scale its funding to materially shift the field — and over what time horizon — is the central economic question. We do not pretend to have solved it. A complementary funding source, governed in the open, may itself be a necessary condition for closing the gap, but the trajectory depends on ecosystem adoption we cannot predict.

3.2 Governance Design for AGI Funding

Designing governance for a DAO that funds AGI research surfaces hard, interlocking questions for which the existing DAO ecosystem has no proven answers:

Quorum and turnout. Real DAOs routinely see turnout below 5%; attacker coalitions thrive in apathy. What participation model works for technical AGI funding decisions, which require informed voters?

Vote weighting. Should voting be 1-token-1-vote, vote-escrow weighted, quadratic, or reputation-based? Each model has known failure modes; none has been stress-tested at AGI-funding scale.

Speed vs. safety. Frontier AGI moves quickly. Deliberate slowness protects against capture, but a DAO that cannot move tactically will be irrelevant. Where the boundary sits is unresolved.

Capture resistance. Governance attacks (vote buying, flash-loan delegation, social engineering) are now well-documented. The defenses are partially understood but unproven in combination.

3.3 AGI Safety Verification

The mission is AGI as an open resource for all of humanity, pursued safely. Verification of deliverables (does the code pass tests, were the weights shipped, was the dataset published) is conceptually distinct from verification of safety (does this output advance or set back alignment). For code and model grants, deliverable verification is tractable; for alignment and capability research, no agreed framework exists. How DAOAGI evaluates funded work for safety impact — and who is qualified to make that judgment — is an open problem we explicitly defer to community design.

3.4 Token Economic Sustainability

The DAO intends to capture a fraction of the value it creates (through IP-NFTs, service fees, bonding-curve fees, partnerships). Whether these mechanisms can plausibly fund a sustained grant program at meaningful scale is unproven. The honest answer is that early rounds will rely on donations and treasury reserves, with value-capture mechanisms maturing over time. The transition path from donated capital to self-sustaining revenue is itself an open design question.

3.5 Legal & Regulatory Uncertainty

DAOs operate in a moving regulatory landscape. Recent case law (CFTC v. Ooki DAO, 2022; Samuels v. Lido, 2024) has shown that unwrapped DAOs can expose participants to personal liability. We believe a dual legal structure (a Swiss Association or Cayman Foundation paired with a Wyoming DUNA or equivalent) is the right direction, but the details must be finalized with DAO-experienced legal counsel in light of the regulatory environment at the time of formation. Securities-law treatment of any governance token is likewise jurisdiction-dependent and must be analyzed before launch.

3.6 Resilience and Recovery

The DAO ecosystem has accumulated a long graveyard of failures: governance captures, smart-contract exploits, vote-escrow model collapses, toxic-culture implosions, and unrecoverable hacks. We have begun designing a defense-in-depth architecture (slow governance, an autonomous immune system, quarantine and recovery modes, an AI agent execution layer) but these mechanisms are unproven at the scale and adversarial pressure an AGI-funding DAO will face. The architecture is a starting hypothesis, not a guarantee.

3.7 Decentralization Trade-offs

DAOAGI will launch with a Founding Council that holds meaningful operational control. This is necessary for speed and safety in the early stages, but it is not full decentralization. The path from council stewardship to genuine community control — and the conditions under which that transition occurs — must be designed transparently and reviewed under real operating conditions.

3.8 AGI-Specific Existential Risks

Funding open AGI research, even with the best alignment intentions, could contribute to capability gains that outpace safety progress. A funding DAO must therefore think about its own outputs as potential contributors to catastrophic risk, not just as public goods. How to weigh this honestly — what to fund, what to delay, what to refuse — is among the hardest questions the community will face.

4. Solution Direction

We propose to explore, with the ecosystem, a funding protocol organized around the following components. The framing here is directional; specific parameters, contracts, and policies will be defined in subsequent versions of this whitepaper through open design work.

Governance token (DAGI). A fixed-supply, pure-governance token that coordinates decisions over treasury allocation, grant approval, and protocol parameters. The token carries no profit expectation.

Community-governed treasury. A smart-contract-controlled treasury holding stablecoins, ETH, and other assets, deployed under explicit risk policies and migrated from council custody to full DAO control as the ecosystem matures.

Milestone-based grant lifecycle. Grants structured as discrete, verifiable milestones; funds released only when milestones are independently confirmed. Each milestone is an open deliverable — software, weights, or research entering the public record. Streaming payments and clawback mechanisms will be explored to align incentives around measurable progress.

Independent verification layer. A combination of automated checks, elected domain-expert reviewers with reputation tied to on-chain attestations, and a dispute-resolution path. Reviewers confirm both that the work was delivered and that it was released as an open resource. The hard problem of safety verification (as distinct from deliverable verification) is explicitly flagged for community design (§3.3).

Value-capture subsystem. IP-NFTs and attribution instruments; optional voluntary service fees; bonding-curve commons fees; partnership and compute revenue; and quadratic-funding donations with Sybil resistance. These instruments exist to fund the mission. They must not restrict anyone’s right to use, study, modify, or share the software, weights, papers, data, or methods produced with DAOAGI funds. The exact mix and parameters are open.

Irrevocable openness commitment. All funded outputs — software, model weights, papers, data, methods, and the knowledge they contain — are released as a public resource. Software ships under an irrevocable open-source license (Apache 2.0 by default). Models ship with open weights. Research ships in full. There is no delayed-open or source-available exception, including Business Source License, for these outputs. What DAOAGI chooses not to fund for safety reasons is a separate question (§3.8); what it does fund is open.

Layered resilience architecture (directional). Slow human governance for strategic decisions, an autonomous immune system for threat response, quarantine and recovery modes for surviving structural crises, and AI agent support for tactical execution and accessibility. These subsystems are designed hypotheses, to be validated iteratively against real operating conditions.

Honest progressive decentralization. Launch under a Founding Council with a published mandate and explicit sunset conditions, transition to full on-chain governance as participation thresholds are met, and publish quarterly decentralization progress reports.

Each component above is a design surface, not a final answer. We expect subsequent versions of this whitepaper to harden each one through open community process. The deeper mechanism work drafted in earlier internal iterations — including the full immune-system tiers, quarantine vaults, recovery doctrine, AI agent defenses, and personal digital twin stack — is preserved as design material to be revisited, stress-tested, and revised with ecosystem input rather than committed to here as final.

5. Guiding Principles

The following principles are intended to outlast any specific mechanism. If a future governance vote must choose between a mechanism and a principle, the principle wins.

Open by requirement. Funded outputs remain a public resource for all of humanity: open-source software, open-weight models, and the full research record. This is not a default with exceptions. Governance cannot vote to close software, weights, research, or knowledge financed by the DAO.

Radical transparency. Treasury flows, grant decisions, reviewer reasoning, and incident reports are public by default.

Progressive decentralization. Centralized functions at launch are explicitly temporary, with named sunset conditions and quarterly public reporting on the transition.

Defense-in-depth resilience. The protocol is designed to respond, protect, recover, and learn from crises — not to assume they will not happen.

Minority protections. Exit and fork rights are constitutionally preserved; no majority can permanently trap a dissenting minority.

Mission over profit. DAGI is a coordination tool, not an investment. Value capture exists to fund the mission, not to enrich token holders.

User sovereignty. Any AI agent or digital twin operating on a user’s behalf remains under that user’s control and consent.

6. Iteration Roadmap

The roadmap below describes a co-design process, not a feature delivery schedule. Dates are indicative; mechanisms ship when the community has hardened them.

Each phase concludes with a public retrospective and an updated version of this document.

7. Disclaimers & Call to Contribute

This whitepaper is a founding statement, not a specification. It is for informational and educational purposes only and does not constitute financial, legal, or investment advice. No governance token has been issued; none will be issued until the legal and regulatory analysis is complete and the community has ratified the design. Participation in DAOAGI, once launched, involves risks including loss of funds, regulatory changes, smart-contract failures, and project failure.

We are publishing early because the problems DAOAGI is taking on cannot be solved by a small group in isolation. If you are a researcher, a builder, a funder, a reviewer, a security engineer, a lawyer, an ethicist, or simply a person who cares that AGI remain an open resource for the entire human society, the design work is open.

Help us fund AGI as an open resource for everyone. The conversation starts here.