OpenAI Chief Executive Officer Sam Altman confirmed Thursday that the company implemented significant changes to its artificial intelligence models following a direct review by Trump administration officials. The revisions allowed for the public release of GPT-5.6, which had been restricted to select partners under pressure from the federal government.
In an interview with CNBC, Altman described the pre-release period as a “collaborative back and forth” with Commerce Secretary Howard Lutnick and Treasury Secretary Scott Bessent. The executive declined to specify the nature of the alterations, but stated the government’s vetting process involved testing new models and identifying potential security vulnerabilities.
Federal Gatekeeping for Frontier Models
The government’s intervention aligns with a broader push to assert American primacy over advanced technology infrastructure. Rival firm Anthropic PBC faced a temporary forced shutdown of its Fable 5 and Mythos 5 models when the Commerce Department imposed export controls designed to block foreign access, a move the company only resolved last month after implementing additional cybersecurity safeguards.
Altman expressed cautious support for the new oversight regime, provided it remains efficient. “I think that’s good as long as the process is understandable, fair and quick,” he said. “The government seems to really share those goals.” This top-down review model suggests the administration views frontier AI not merely as a commercial product but as a strategic national asset requiring protection from foreign adversaries.
Equity Rumors and Market Position
Altman pushed back against recent reports alleging OpenAI proposed giving the U.S. government a five percent ownership stake, dismissing the claims as containing “a lot of inaccuracies.” When pressed on the likelihood of an initial public offering this year, Altman offered only a succinct “I don’t know,” keeping the company’s financial trajectory opaque as it navigates the new regulatory landscape.
The episode underscores a growing reality for the domestic AI industry: access to the American consumer market now requires navigating a federal gatekeeping apparatus that prioritizes national security over rapid, unbridled deployment.