Trump administration seeks voluntary AI model reviews from Meta

Trump administration pressures Meta on AI security reviews

The Trump administration has asked Meta to submit its artificial intelligence models for voluntary government review, making the social media giant the sole major U.S. AI developer that has not yet agreed to federal security assessments.

The request came through confidential emails between administration officials and Meta, according to a New York Times report citing four people familiar with the communications. The voluntary reviews would enable government officials to evaluate the capabilities and security vulnerabilities of Meta's AI systems.

Meta responded that it shares the administration's goal of advancing U.S. leadership in secure frontier AI development. "While we are working through the details, we hope to sign the agreement soon," a Meta spokesperson told Reuters.

The Commerce Department did not immediately respond to requests for comment on the matter.

Why it matters

The push for voluntary AI reviews reflects growing federal concern about national security risks in advanced AI systems, particularly around potential vulnerabilities that foreign adversaries could exploit. Meta's open-source approach to AI development—releasing powerful models like Llama publicly—has made it an outlier among major tech companies, raising distinct security questions for regulators. The company's holdout status on voluntary reviews could signal friction over how open-source AI fits into emerging government oversight frameworks.

Broader AI oversight intensifies

The request to Meta comes as the administration escalates scrutiny across the AI sector. Earlier in June, the U.S. government ordered Anthropic to suspend foreign nationals' access to its most advanced AI models, citing national security concerns.

According to the Times report, Meta is now the only major U.S. AI technology developer that has not reached an agreement to voluntarily share its models with the federal government for security review. Other leading AI companies, including OpenAI, Google, and Microsoft, have already established frameworks for government assessment of their systems.

The voluntary nature of these reviews distinguishes them from mandatory regulatory requirements, though the administration's public pressure on Meta suggests the line between voluntary and compulsory cooperation may be blurring as AI capabilities advance.

Meta has pursued a notably different strategy from competitors by releasing many of its AI models as open source, making them freely available for download and modification. This approach has drawn both praise for democratizing AI access and criticism over potential security implications.

The details were first reported by the New York Times.