Senior Trump administration officials had weighed how to structure potential government equity stakes in major AI companies before the government’s export controls on Anthropic further roiled the industry.

Two top Cabinet members had discussed different ideas, people familiar with the talks told Semafor: Treasury Secretary Scott Bessent favored using equity in AI firms to seed Trump Accounts, while Commerce Secretary Howard Lutnick’s preference was that any equity be directed to a type of sovereign wealth fund.

The talks about possible AI stakes ceded to the government are still in the early stages, with no decision made yet — and a meeting with industry CEOs that President Donald Trump previewed earlier this month yet to emerge.

Given that, it’s still unclear where the Trump administration could ultimately land on an idea that remains an extremely tough sell for most of the industry beyond OpenAI, which first pitched it last year. Executives at firms like Microsoft and Meta have turned a cold shoulder in the last week alone.

The administration’s crackdown on the latest models from Anthropic may also complicate equity talks. Trump told reporters last week that he would convene “the top 12 or 15 executives very shortly” to discuss the AI industry “giving back something to the public,” but Friday’s export controls risk turning any meeting into a more tense affair.

“What AI companies at all levels and structure need is proper guardrails and federal legislation on how we’re going to use artificial intelligence in the government,” Caleb Max, president and CEO of the National Artificial Intelligence Association, said Friday before the Anthropic news broke. “I don’t think it’s infusing cash in them at this point.”

The meteoric success of SpaceX’s IPO recently set the stage for subsequent public offerings from OpenAI and Anthropic, which both could see valuations in the $1 trillion range. That raises further questions for some in Washington about whether industry competitors will ever wind up ceding equity to the administration.

Sen. Cynthia Lummis, R-Wyo., told Semafor the proposal was a “head-scratcher,” adding: “It starts out with my nose wrinkled.” At a recent Senate Banking Committee hearing, Sen. John Kennedy, R-La., also raised concerns.

“I don’t mean disrespect when I say this — I’m proud of them, as American entrepreneurs — but I trust them like I trust the rest-stop bathroom,” Kennedy said of AI company leaders.

One AI lobbyist expressed doubt that any AI company — even those who have paid lip service to the idea — would make any real effort to bring it to fruition.

“You do a little bit of a pressure test based on circumstance or whatever and you see that it all falls apart,” the lobbyist said.

The White House, Treasury, and Commerce did not comment for this story, nor did OpenAI or Anthropic.