Sam Altman wants to sell you a story. The OpenAI CEO has floated a proposal to hand the U.S. government a 5% equity stake in his company, worth roughly $42.6 billion at OpenAI’s $852 billion March 2026 valuation as part of a “Public Wealth Fund” modeled on the Alaska Permanent Fund. On its face, it sounds almost noble. Let the American people own a piece of the AI revolution, get dividends, share in the upside. But peel back the press-release veneer and what you find is something far uglier. A regulatory protection racket dressed up in progressive clothing. This isn’t wealth-sharing. This is pay to play.
The $42.6 Billion Insurance Policy
OpenAI closed a $122 billion funding round in March 2026 at an $852 billion valuation, backed by Amazon, Nvidia, SoftBank, and Microsoft. The company is now generating $25 billion in annualized revenue and has filed a confidential S-1 with the SEC for an IPO that could push its valuation past $1 trillion. A 5% government stake at current valuation is $42.6 billion. At a trillion-dollar IPO, it’s $50 billion. That’s a lot of money to hand over unless you’re getting something even more valuable in return.
What OpenAI gets is regulatory immunity. By making the government a shareholder, Altman creates a structural conflict of interest so profound it makes the revolving door between Wall Street and Washington look like child’s play. The U.S. government would simultaneously be the regulator of AI safety and a financial stakeholder in the largest AI company on the planet. Every safety rule, every antitrust action, every export control decision that could ding OpenAI’s stock price would face an implicit question: will this hurt our portfolio?
As governance experts from Public Knowledge have warned, this arrangement creates a “substantial conflict of interest” that could compromise the government’s ability to regulate AI in the public interest. The government would have a fiscal incentive to go easy on OpenAI, to look the other way on safety violations, to steer procurement contracts its way, to slow-walk competitor approvals. This isn’t conspiracy theory. This is basic incentive structure.
The Alaska Permanent Fund Analogy is a Fraud
Altman loves to invoke the Alaska Permanent Fund as his model. Here’s how it works: Alaska takes 25% of its oil revenue, invests it in a diversified portfolio, and pays an annual dividend to every resident. In 2024, that dividend was about $1,702 per person. It’s a genuinely popular program that has made Alaska one of the most income-equal states in America. But the analogy is intellectually dishonest in three critical ways.
First, the Alaska Permanent Fund is funded by a tax on resource extraction. The state takes a cut of oil companies’ revenue because the oil belongs to the people of Alaska. OpenAI’s proposal is the opposite: the company is voluntarily offering equity to buy goodwill, not because the public has any ownership claim on its technology. There’s no tax, no legal mandate, no democratic process. It’s a bribe with better branding.
Second, the Alaska Fund is managed by an independent public corporation with a constitutional mandate to preserve principal and distribute earnings. Altman’s proposal has no such guardrails. Who votes those shares? Does the government get board seats? Can the administration of the day use those shares to influence OpenAI’s business decisions? The answers are: nobody knows, probably not, and absolutely yes.
Third, the Alaska model distributes cash directly to citizens. OpenAI’s proposal doesn’t specify any distribution mechanism at all. It’s a “wealth fund” in name only. It is more likely a slush fund that politicians can raid for pet projects while claiming they’re “investing in the American people.”
The Safety Exodus Makes This Even Worse
Here’s the part Altman really doesn’t want you to think about. While he’s busy negotiating equity stakes with the Trump administration, his safety team is evaporating.
The list of departures reads like a who’s-who of AI alignment research: co-founder Ilya Sutskever (May 2024), Superalignment co-lead Jan Leike (May 2024, who publicly said “safety culture and processes have taken a backseat to shiny products”), AGI Readiness lead Miles Brundage (October 2024), VP of Research and Safety Lilian Weng (November 2024), Preparedness head Aleksander Madry (May 2026), Mission Alignment lead Joshua Achiam (July 2026), Safety Systems head Johannes Heidecke (July 2026). The list goes on.
OpenAI has repeatedly dissolved its independent safety teams, folding them into the research division under the VP of Research and Safety, the same person responsible for shipping products. Critics call this what it is: structural capture of safety by commercial incentives.
Now ask yourself: if the U.S. government owns 5% of OpenAI, and the company’s own safety researchers are fleeing because safety keeps losing to “shiny products,” what incentive does a shareholder-government have to demand stronger safety protocols? The answer is none. A government that owns OpenAI stock benefits from faster releases, higher valuations, and fewer regulatory delays, the exact opposite of what responsible AI governance requires.
The Sanders Alternative is Worse (But Honest)
Senator Bernie Sanders has introduced the American AI Sovereign Wealth Fund Act, which would impose a one-time 50% tax on the equity of major AI companies including OpenAI, Anthropic, Google, and Meta to fund a public wealth fund estimated at $7 trillion. The bill would give the government voting shares and board representation, managed by an Independent Commission for Democratic AI.
Sanders’ proposal is radical, probably unconstitutional, and politically dead on arrival in a Republican-controlled Congress. But at least it’s honest. It doesn’t pretend the government can regulate an industry it owns nor doesn’t pretend the arrangement is voluntary. It says: AI was built on the collective knowledge of humanity, and the public deserves a real seat at the table.
Altman’s 5% offer is designed to preempt exactly this kind of legislation. It’s a small, voluntary concession that makes the larger, democratic alternative seem unnecessary. It’s the oldest trick in the corporate playbook: offer just enough to kill the real reform.
OpenAI offered the US government a 5% stake — $42.6B — modeled after Alaska's oil fund. "AI wealth belongs to the public," Altman said. Timing: right before OpenAI's IPO, while regulators decide its fate.#OpenAI #AI #AIPolicy pic.twitter.com/tmGGKreTGV
— AEVORA (@zflippers) July 14, 2026
OpenAI Reportedly Secures US Government Clearance to Launch GPT-5.6 Model
— Cyber Security News (@The_Cyber_News) July 8, 2026
Source: https://t.co/pWv2ZyXIy5
OpenAI has reportedly received approval from the U.S. Department of Commerce for a broad public launch of its advanced GPT-5.6 model, marking a significant moment in how… pic.twitter.com/fbQ2zqj0XC
The Governance Guardrails That Don’t Exist
Let’s be concrete about what a responsible government equity arrangement would require:
- Blind trust management: The government’s shares must be held in a blind trust with no communication between the trust manager and regulators. No one at the FTC, DOJ, or Commerce Department should know when the government is buying or selling.
- No voting rights: The government should have zero ability to influence corporate governance through its shares. No board seats, no shareholder votes, no say in executive compensation.
- Mandatory distribution: All dividends or proceeds must be distributed directly to citizens as cash payments, not funneled into government programs or slush funds.
- Sunset clause: The arrangement must have a fixed term, subject to renewal only by an act of Congress with supermajority approval.
- Independent audit: An independent ethics office must publish annual reports on every regulatory decision involving OpenAI, with a public finding on whether the government’s stake created any conflict.
None of these guardrails exist in Altman’s proposal as of right now. The plan is conceptual, which in Washington-speak means “we’ll figure out the details after you’ve already agreed to the principle.” And once the principle is agreed, the details will be written by the people who stand to benefit.
In Bed With The Treasury Department
OpenAI’s 5% government equity proposal is not a public wealth fund. It’s a $42.6 billion regulatory capture fee, a down payment on a future where the U.S. government has a financial stake in looking the other way while the most powerful technology in human history is developed behind closed doors by a company that has already demonstrated it cannot be trusted to prioritize safety over speed.
The Alaska Permanent Fund works because it’s funded by a tax on extraction, managed independently, and distributed directly to citizens. Altman’s proposal has none of those features. It’s a simulacrum of wealth-sharing designed to neutralize political opposition while preserving every ounce of corporate control. The moment the Treasury Department starts worrying about OpenAI’s stock price is the moment AI safety becomes a secondary concern.
Altman is offering the American people a seat at the table. But the table is his, the rules are his, and the only thing on the menu is whatever he decides to serve. That’s not partnership. That’s pay to play and we’re the ones paying.
Leave A Comment