A new report discussed in The New York Times DealBook newsletter reveals that the Trump administration is considering a major shift in its approach to artificial intelligence regulation. After initially promoting a largely hands-off strategy designed to accelerate American AI dominance, officials are now reportedly discussing government oversight mechanisms for advanced AI systems before they are publicly released. The move reflects growing concern in Washington about the national security, cybersecurity, and societal risks posed by increasingly powerful AI models.
According to reports, the White House is weighing an executive order that could create a formal review process involving government officials and technology executives. Discussions intensified following concerns around highly capable AI systems such as Anthropic’s Mythos model, which cybersecurity experts warned could identify and exploit vulnerabilities in software infrastructure at alarming speed. Officials are also reportedly developing an AI security framework that would require safety testing for AI systems used across federal, state, and local governments.
The potential policy reversal is especially notable because the Trump administration previously rolled back several Biden-era AI safety initiatives and pushed for limits on state-level AI regulation. Earlier executive actions emphasized reducing barriers to innovation and preventing what officials described as a fragmented regulatory environment across U.S. states. However, rapid advances in autonomous AI capabilities, cybersecurity risks, and geopolitical competition with China appear to be forcing a more interventionist stance from policymakers.
The debate has divided experts and industry leaders. Supporters of stronger oversight argue that frontier AI models now pose risks significant enough to justify national review systems similar to those used in other high-risk industries. Critics, meanwhile, warn that excessive regulation could slow innovation, consolidate power among large tech firms, and weaken America’s competitive position in the global AI race. The discussion highlights how governments worldwide are struggling to balance innovation, economic leadership, and safety as AI technology evolves faster than existing regulatory systems.
