The White House AI regulation push targets top tech leaders to finalize safety guidelines and mandatory model testing.
White House officials are set to meet with leaders from top AI companies on Tuesday to discuss a new pre-launch government review process for frontier AI models, sources told CNN.
The meeting represents a major step toward broader AI regulation as both tech firms and lawmakers push for tighter controls on AI development. While the review framework stems from an executive order signed by President Donald Trump in June, the talks follow recent reports from OpenAI and Anthropic regarding AI agents rogue-hacking external systems.
Under the executive order, the new system gives the government access to the most advanced AI models up to 30 days before public release. Participation remains voluntary, though the administration has recently acted to delay or block certain models over security concerns.
“The voluntary framework outlined in the June 2 executive order is complete. Discussions with industry about next steps are underway,” a White House official told CNN on Monday.
Tuesday’s meeting will include OpenAI, Anthropic, Google, and Meta, according to sources familiar with the matter. While Anthropic, Google, and Meta declined to comment, OpenAI pointed to a blog post by Chief Global Affairs Officer Chris Lehane reiterating calls for Congress to establish national AI standards.
“The Administration’s expected action this week on frontier AI could be an important step toward closing the gap between innovation and governance,” Lehane wrote, calling for a clear national framework to evaluate top-tier AI models quickly and safely.
However, the details and standards of the framework will remain hidden from the public, with many classified under the June executive order.
While major AI developers worked closely with the White House on the executive order and framework, sources familiar with the talks say significant uncertainty remains over the final details.
Key unresolved questions include how the administration will define “frontier AI models” and whether open-weight models which can be downloaded locally and modified will fall under the review process.
It also remains unclear which government body will lead the review process. According to sources within the participating tech firms, the White House has yet to designate a single point person or agency to direct industry outreach.
For now, the effort has been spearheaded by National Cyber Director Sean Cairncross, Treasury Secretary Scott Bessent, and Commerce Secretary Howard Lutnick. However, administration officials note that full oversight will require cooperation across multiple government branches given AI’s broad impact.
Adding to the uncertainty is friction between the administration and specific AI developers most notably Anthropic. The Pentagon designated Anthropic a “supply chain risk” earlier this year following a dispute over safety guardrails in its models, which the military had sought to deploy.
See also: NGX Domestic Stock Trading N17.7T in 18 Months
The blacklisting an unprecedented action against an American firm prohibits the military and its partner organizations from using Anthropic’s products.
While Anthropic and the Pentagon contest the designation in court, the White House has maintained a productive dialogue with the company on the executive order and its newest model releases. President Trump even told Axios in June that he no longer considers Anthropic a national security threat, praising the startup for acting “very responsibly.”
Pressure for AI regulation has mounted following revelations that AI agents from OpenAI and Anthropic independently hacked external systems during routine cybersecurity tests. Last week, over 1,200 senior staff across top AI firms including Anthropic CEO Dario Amodei signed an open letter urging the U.S. government to develop tools that intentionally slow advanced AI development until safety measures can catch up.
“We may have to pace the rate of AI development to give ourselves enough time for society to harden around these new capability levels,” OpenAI CEO Sam Altman noted on a recent episode of the Invest Like the Best podcast.
Meanwhile, the European Union’s AI Act took effect on August 2, empowering the European Commission to review advanced models prior to release. The law mandates strict transparency and documentation standards, threatening fines of up to 7% of a company’s annual gross revenue for non-compliance.
