US President Donald Trump has signed a new executive order aimed at strengthening cybersecurity safeguards around advanced artificial intelligence systems, creating a framework for the government to assess potential national security risks before cutting-edge AI models are released to the public.
The move comes amid growing concerns over the capabilities of increasingly powerful AI systems and follows recent debate within the technology industry over how such models could be exploited for cyberattacks or other malicious purposes.
The executive order establishes a voluntary partnership between the federal government and major AI developers, including OpenAI, Anthropic, Microsoft, Google and other leading technology companies. Under the framework, participating firms may provide government experts with access to their most advanced AI models up to 30 days before public launch.
According to the White House, the goal is to allow federal agencies to evaluate whether frontier AI systems pose cybersecurity or national security risks and to identify potential vulnerabilities before they become widely available.
The administration said the programme would include confidentiality protections designed to safeguard companies’ proprietary technology and trade secrets.
A key focus of the order is the growing threat of AI-enabled cyberattacks. The directive instructs federal agencies to work closely with AI developers to strengthen defences against malicious uses of artificial intelligence, including sophisticated hacking campaigns, automated cyber intrusions and threats targeting critical infrastructure.
The order also seeks to expand access to advanced cybersecurity tools powered by AI. Federal, state and local government agencies, along with operators of critical infrastructure such as energy grids, transportation networks and communications systems, would be encouraged to use security capabilities embedded within frontier AI models.
Supporters of the measure argue that closer cooperation between government and industry is necessary as AI systems become increasingly capable and potentially disruptive. They contend that early evaluation of advanced models could help identify dangerous capabilities before they are exploited by hostile actors.
The announcement comes only days after Trump warned that excessive regulation of artificial intelligence could weaken the competitive position of US technology companies against rivals, particularly China. The new order attempts to balance those concerns by relying on voluntary participation rather than imposing mandatory pre-release reviews or regulatory restrictions.
As AI development accelerates, governments around the world are grappling with how to manage the technology’s benefits while reducing risks related to cybersecurity, national security and public safety. The latest executive order represents the Trump administration’s effort to address those concerns through collaboration with the private sector rather than formal regulation.
