President Donald Trump signed a revised executive order on AI after senior aides persuaded him to finalize a framework for the technology, according to two officials. The order allows the federal government access to the most advanced AI models 30 days before their public release, a reduction from the initial 90-day proposal. The revised order reflects growing concern within the administration about the potential for powerful systems, including Anthropic's Claude Mythos and OpenAI's GPT-5.5, to be exploited for cyberattacks against critical infrastructure. The executive order is the first major AI regulation directive of Trump's second term. Source: wired
The revised order establishes a voluntary process to identify the most powerful AI models and provides the US government exclusive access for 30 days to address potential vulnerabilities before public release. The move marks a victory for White House Chief of Staff Susie Wiles, who worked with Treasury Secretary Scott Bessent and National Cyber Director Sean Cairncross to revive the proposal. Despite initial resistance from former AI czar David Sacks, the administration proceeded with the plan. The order also enables discussions with China on creating a cross-border framework for advanced AI systems, according to a person familiar with the matter. Source: wired
The executive order directs federal agencies to create a classified process to determine which AI models the US government wants access to and select trusted partners for early access. It also mandates the Pentagon to strengthen its classified networks within 30 days and the Justice Department to pursue criminal cases against individuals using AI models to hack computer systems. White House spokesperson Liz Huston emphasized the order as a reflection of Trump's 'common-sense approach' to balance innovation and security. Source: wired
