Multiple reports say the Trump administration is urging Meta to allow the U.S. government to review its most capable AI models for federal security purposes. According to a New York Times report cited by both outlets, Meta is the last major U.S. AI developer that has not agreed to participate in these reviews. The pressure reportedly comes through email communications between government officials and Meta, as Washington increases oversight of frontier AI systems. The reports frame the effort as part of a broader push to evaluate higher-capability models that could affect national security or other sensitive areas. While details about the exact review process and what models would be included are not specified in the excerpts, the reporting is consistent that the administration’s requests center on Meta submitting its leading AI models for evaluation. Meta’s reported position is described as holding out, in contrast with other major developers that have reportedly already agreed to provide models for federal security review. The situation reflects an ongoing U.S. policy push to expand government access to advanced AI systems.