OpenAI has confirmed it will participate in the voluntary framework established by President Donald Trump's executive order asking AI companies to give the federal government access to their most powerful models before public release.
George Osborne, who leads OpenAI's country-level operations, confirmed that the ChatGPT maker would comply with the order, according to CNBC. "It's quite right that democratic governments have a big role to play in how this technology is used and deployed," he said. Osborne previously served as the U.K.'s finance minister from 2010 to 2016.
At SXSW London, Osborne made clear that OpenAI acts without waiting for government prompts, describing the company's sense of obligation as something it takes "very seriously." He added that OpenAI has put forward its own proposals to help governments monitor safety and security risks tied to AI — an effort he said extends beyond the United States.
Trump signed the order earlier this week. Participating companies would be required to give federal agencies a window of up to 30 days to conduct cybersecurity evaluations of their top-tier models ahead of any broader rollout. Whether a model qualifies as a "covered frontier model" — the threshold that triggers the early-access provision — would be determined through a classified benchmarking assessment of its advanced cyber capabilities. The order explicitly prohibits any reading of it as authorizing mandatory licensing, preclearance, or permitting requirements for AI model development or release.
OpenAI CEO Sam Altman backed the order after its signing, writing on X $TWTR that it "gets the balance right" and that the U.S. should lead on AI "by continuing to develop the very best models, making sure they're safe, and getting cyber tools into the hands of trusted defenders." Altman subsequently met with White House officials and members of Congress, including House Speaker Mike Johnson and House Minority Leader Hakeem Jeffries.
The order went through months of deliberation at the White House, including a last-minute pull by Trump in May over concerns that certain provisions could undermine U.S. competitiveness against China. Its final shape was influenced in part by the emergence of Anthropic's Mythos model, which drew attention in Washington for its ability to identify vulnerabilities in widely used software. OpenAI, Google $GOOGL, and Anthropic all met with federal officials as the policy was being developed.