OpenAI offers Washington $42 billion to head off a Senate bill taxing AI at 50%

OpenAI CEO Sam Altman has put a specific number on the idea that the public should share in AI’s financial upside: $42.6 billion. That is what a 5% ownership stake in OpenAI is worth at the company’s current $852 billion valuation, and Altman has proposed handing it directly to the US government through a sovereign wealth fund modeled on Alaska’s Permanent Fund — the state-owned vehicle that has paid annual oil dividends to Alaskans since 1976.

The proposal, first reported by the Financial Times on July 2 and confirmed by Bloomberg and CNBC, involves talks between Altman and President Trump, Commerce Secretary Howard Lutnick, and Treasury Secretary Scott Bessent. Altman’s pitch does not stop at OpenAI: he wants every leading US AI company — Anthropic, Google, Meta, and xAI — to contribute matching 5% stakes to the same fund.

The timing is not accidental. Senator Bernie Sanders has proposed the American AI Sovereign Wealth Fund Act, which would impose a 50% tax on major AI companies. The OpenAI proposal is a calculated alternative: give Washington voluntary equity now, and make the argument that a 5% stake is a better deal for the public than a tax that would pull cash out of companies still in their growth phase.

The model works best if the whole industry joins. A fund that holds only OpenAI’s equity is a boutique investment; a fund that holds equity in OpenAI, Google DeepMind, Anthropic, Meta AI, and xAI starts to look like the public receiving a dividend on AI broadly. Alaska paid each eligible resident $1,000 from its fund last year — that is the outcome Altman is pointing at.

As of this week, no competing AI company had agreed to participate. Neither the White House, Google, Meta, nor Anthropic responded to press inquiries. Discussions are described as “very early” and “conceptual.” Forrester analyst Indranil Bandyopadhyay warned that a US model could trigger other governments to demand equivalent market-access conditions — a potential legal and trade complication for AI companies operating globally.

A government stakeholder would also have an inherent interest in the company’s profitability, which may not align with aggressive AI safety oversight. For users outside the United States — the majority of people who use ChatGPT — the proposal is structurally irrelevant: a US sovereign wealth fund funds American programs only, not global users whose interactions trained the models. Congressional discussions are expected into fall 2026.

Tags: artificial intelligence, OpenAI, Sam Altman, AI policy, sovereign wealth fund