OpenAI has committed to allowing US government regulators to review its artificial intelligence models before public release. The decision comes in response to President Trump’s recent executive order aimed at ensuring AI safety through government oversight.
The move signals a significant shift in how major AI companies will work with federal authorities, potentially setting a precedent for the entire industry as concerns about AI safety continue to grow.
OpenAI announced its compliance with the order this week, with the company’s head of countries George Osborne telling CNBC that “democratic governments have a big role to play in how this technology is used and deployed.” He suggested that governments should “create powerful regulatory bodies, but with a lot of flexibility into how they will operate in the future.”
Trump’s executive order emerged after considerable back-and-forth between the administration and tech industry leaders. The original proposal would have required companies to submit models 90 days before public release, but industry figures including David Sacks and Elon Musk reportedly warned this could create a “chilling effect” on AI development.
The final version represents a compromise that reduces the review period to 30 days and makes participation voluntary rather than mandatory. The order requests that AI companies participate in a benchmarking process to evaluate advanced cyber capabilities of their models and determine whether they qualify as “covered frontier models” – a designation that could restrict their distribution and sale.
The policy change reflects growing government attention to AI safety as models become more powerful and widespread. Federal authorities want to identify potentially dangerous capabilities before they reach the public, particularly in areas like cybersecurity where AI could be used maliciously.
However, the voluntary nature of the order has drawn criticism from lawmakers who believe stronger measures are needed. Rep. Don Beyer (D-VA), who co-leads an AI-focused congressional group, called the policy “underwhelming” and said it creates “a wild west environment for AI development.”
For OpenAI, cooperating with government review could help the company maintain its position as a responsible AI leader while potentially influencing how future regulations are shaped. The company has previously called for AI regulation and safety standards, making this compliance consistent with its public stance.
The executive order’s impact will likely depend on how many other major AI companies follow OpenAI’s lead. With participation being voluntary, the effectiveness of government oversight will hinge on industry cooperation and whether competitive pressures encourage or discourage compliance.
