
From theory to table stakes
The AI-government equity-stake idea has officially left the group chat. On June 10, Trump said he plans to meet with AI executives soon, and the conversation is expected to include giving the public a share of the industry’s wealth. In other words: the administration isn’t just talking about regulation anymore, it’s flirting with ownership.
Intel is the giant neon sign
Intel is the blueprint here. In August 2025, the U.S. government converted CHIPS Act funding into 433.3 million INTC shares, landing a 10% non-voting stake at $20.47 a pop. That matters because it shows the playbook: if Washington already has leverage, it may try to turn that leverage into equity instead of handing out a plain-vanilla subsidy like candy at Halloween.
Why AI companies are sweating
OpenAI and Anthropic are now in the blast radius. OpenAI filed a confidential draft S-1 on June 8 and is reportedly aiming for a September listing, while Anthropic filed its own confidential S-1 on June 1 after a monster funding round. If the government starts asking for pre-IPO shares — even a 1% to 5% “donation” — that changes the math for every future retail investor buying into these names.
- A 1% stake in OpenAI at roughly an $852 billion valuation is still north of $8.5 billion.
- A voluntary deal could be the “cheaper” option versus a harsher bill later.
- But once Washington is on the cap table, governance gets weird fast.
Big picture
The real investor takeaway is that AI IPOs may not just be about growth anymore — they may come with political strings attached. If the Intel model becomes the template, the cost may be manageable. If it evolves into something closer to a mandated grab, the whole sector’s public-market coming-out party gets a lot less festive.
