Last week, Public First Action dropped $15 million on political ads targeting AI safety. The numbers are stark: 16 Republican lawmakers, 700,000+ deployed spots across key districts. But this isn't just another political spend—it's a canary in the coal mine for how centralized power shapes the future of artificial intelligence. And it's exactly why blockchain's lessons on governance have never been more urgent.
We don't need permission to see the pattern. The same concentration of capital that drove 80% of ICO value to insiders in 2017 is now buying influence over the rules that will govern the most transformative technology since the internet. As someone who built communities through the DeFi summer and watched governance tokens get captured by whales, I recognize the smell: a small group of undisclosed donors using a Super PAC to steer AI regulation toward their preferred outcome. The goal? Make AI safety a partisan wedge issue, secure friendly lawmakers, and lock in a regulatory framework that benefits the incumbents.
Context: The AI Safety Political Divide
Public First Action is a Super PAC—a political action committee that can raise unlimited funds from corporations, unions, and individuals, as long as it doesn't coordinate directly with campaigns. Its stated mission: “support candidates who prioritize AI safety.” The $15 million commitment is significant not just for its size—roughly equivalent to the entire 2024 budget of the Federal Election Commission's enforcement division—but for its timing. We're in the middle of the 2026 midterm cycle, with control of Congress hanging in the balance.
Inside the Republican party, a split is forming. One camp, the “safety hawks,” argues that AI poses existential risks—electoral deepfakes, bioweapons, autonomous warfare—and demands strict federal oversight. The other, the “free-market faction,” insists that regulation will stifle innovation and cede dominance to China. The PAC's money is designed to tip the scales: reward incumbents who back safety bills, and fund primary challengers against those who don't.
Freedom isn't free, but it's also not bought with TV ads. The irony is that both sides claim to defend liberty, yet neither questions the centralization of decision-making power. The $15 million doesn't come from a grassroots collection of concerned citizens; it comes from a handful of tech billionaires and corporate treasuries whose identities remain hidden. According to FEC filings, the top donors to similar AI-focused PACs in 2024 included founders of major AI labs, venture capital firms, and even some cryptocurrency billionaires. The pattern is clear: those with the deepest pockets get to define what “safe” means.
Core: Centralization Risks in AI Governance
Let's strip away the hype and look at the mechanics. A Super PAC spends money on ads that frame a candidate's stance on AI safety. Those ads run in swing districts, shifting voter perception. The candidate wins or loses based on that perception. Once elected, the candidate votes on bills like the AI Transparency Act, the Algorithmic Accountability Act, or the National AI Commission Act. The entire loop—money, media, legislation—represents a centralized feedback system where influence is proportional to wealth.
Compare that to how decentralized communities handle similar decisions. In my work with the Uniswap and Aave governance forums, I saw firsthand how treasury allocations and protocol upgrades were debated openly on-chain. Every proposal had a transparent trail: who submitted it, how much voting power they wielded, and which delegates supported it. Quadratic voting mechanisms allowed smaller stakeholders to pool their influence. Yes, whale dominance existed—but at least it was visible and auditable. Anyone could fork the code and start a new community if they disagreed.
Based on my experience auditing failed DeFi protocols during the 2022 bear market, I learned that centralization often hides in plain sight. The whitepaper promises decentralization, but the key decision-makers control the multi-sig wallet. Similarly, Public First Action's ads might promote “AI safety,” but the actual regulatory language will be drafted by a handful of staffers who answer to those same donors. The transparency that blockchain brings—immutable records, public ledger, permissionless participation—is precisely what's missing from this political process.
Consider the data: according to OpenSecrets, the top 10 donors to AI-related political spending in 2024 accounted for 78% of all contributions. That's a Gini coefficient of 0.85—higher than Bitcoin's wealth distribution. The same concentration exists in the PAC itself. Without mandatory disclosure of its donors, we can't even assess conflicts of interest. Is an AI lab funding ads that attack a candidate who supports open-source model release? That's a governance attack disguised as public service.
Contrarian: Even Decentralized Governance Can Be Captured
Now let me play devil's advocate. The crypto community often holds up DAOs as the pure alternative to centralized governance. But I've seen DAOs fail too. In 2021, a single whale controlled 60% of the voting power in a prominent lending protocol and single-handedly vetoed a crucial risk parameter update. Decentralization doesn't automatically prevent capture; it just changes the form. Whitepapers become marketing, and governance tokens become tools of influence.
Similarly, Public First Action's strategy may actually be a rational response to a broken system. The US political landscape is already dominated by corporate money; if AI safety proponents don't spend, the other side will. The $15 million might be the only way to counterbalance the lobbying power of tech giants who want zero regulation. In that sense, the PAC is doing what any stakeholder would—using available tools to protect their interests.
But here's the key difference: on-chain governance leaves a forensic trail. Every vote, every delegation, every treasury transfer is recorded forever. Public First Action's ads will disappear after the election; their donor list may never be fully known. The opacity is the problem, not the spending itself. When we say “decentralize AI governance,” we don't mean remove all oversight—we mean make the oversight transparent, accountable, and resistant to capture by any single entity.
s built by our shared vision. The vision for AI governance should be one where decisions are made not by the size of your wallet but by the weight of your arguments. Blockchain offers the technical infrastructure to achieve that: verifiable identity, quadratic voting, retroactive funding, and reputation systems. Projects like Gitcoin and Optimism have already experimented with these mechanisms for allocating public goods. Why not apply them to AI safety?
Takeaway: The Opportunity for Decentralized AI Governance
The Public First Action episode is a warning sign, not a death knell. It shows that the fight for AI's future is being fought with old weapons—money and media—on a battlefield designed by incumbents. But crypto has spent the last decade building new weapons: smart contracts, DAO tooling, zero-knowledge proofs, and decentralized identity. We have the tools to create a parallel governance system for AI that is transparent, global, and permissionless.
Imagine a global AI safety fund, governed by a DAO, where stakeholders from around the world vote on research priorities, red-teaming budgets, and release schedules. Imagine verifiable credentials that allow individuals to prove they are human and have expertise, without relying on corporate gatekeepers. Imagine audits of frontier models published on-chain, with cryptographic proofs of compliance. This isn't science fiction—projects like Worldcoin, Proof of Humanity, and zkML are already building the primitives.
The question is whether we have the will to build it. The $15 million from Public First Action will influence the next twelve months of US AI policy. But the next twelve years of AI governance will be shaped by the infrastructure we lay down today. As a community that understands the cost of centralization—both in financial crashes and lost autonomy—we have a responsibility to offer an alternative.
Volatility is the price of freedom. Stay liquid. But also stay principled. The next time you see a Super PAC ad about AI safety, ask yourself: who paid for this, and what do they want in return? The answer should be on-chain.