US Moves to Regulate AI After Anthropic Security Threat

The Trump administration is developing an executive order to establish a vetting system for advanced artificial intelligence models ^[1]. This move follows reports that Anthropic’s new Mythos model can find and exploit software vulnerabilities on its own ^[1][2]. Officials compare the proposed process to drug testing protocols used by the Food and Drug Administration ^[1].

The Mythos model’s ability to independently target critical infrastructure has caused significant concern within the White House ^[2]. This discovery has reportedly disrupted existing AI strategies, prompting a faster shift toward mandatory safety checks ^[2][3]. The government aims to prevent similar risks before new models reach the public ^[1].

Legal experts are currently weighing how these regulations will impact the tech industry ^[3]. The proposed policy seeks a balance between national security and technological innovation ^[3]. While the executive order is still in preparation, it marks a major change in how the U.S. governs emerging technology ^[1][2].