US President Donald Trump on Tuesday signed a sweeping executive order to accelerate artificial intelligence innovation while fortifying cyber defences across federal systems and critical infrastructure. The directive, issued from Washington, instructs federal agencies to prioritise the protection of national security systems, military information networks and civilian government networks, while scaling the use of AI-powered cybersecurity tools nationwide.

Key Provisions of the Order

According to the White House, the order seeks to ‘advance American artificial intelligence (AI) innovation to strengthen America's cybersecurity, protect critical infrastructure, and ensure the United States remains the global leader in AI innovation.'

The Department of Homeland Security, working with partner agencies, will issue directives to speed up cyber defence efforts and widen access to advanced AI-enabled cybersecurity tools for federal agencies, state and local governments, and operators of critical infrastructure — including rural hospitals, community banks, and local utilities.

AI Cybersecurity Clearinghouse

The order establishes an AI cybersecurity clearinghouse that will work voluntarily with AI companies and infrastructure operators to identify software vulnerabilities, coordinate responses, and accelerate distribution of security patches. Federal agencies have also been directed to identify funding opportunities for advanced AI cybersecurity technologies and expand cybersecurity hiring within the federal workforce.

Benchmarking Frontier AI Models

A central feature is a classified benchmarking process to evaluate advanced AI models for cyber capabilities. The framework will determine when a model qualifies as a ‘covered frontier model' and set up voluntary cooperation between AI developers and the federal government.

The administration said the arrangement would allow trusted government partners early access to advanced AI systems before broader deployment, thereby strengthening cybersecurity and supporting secure innovation. Notably, the order explicitly states that it does not authorise ‘any mandatory governmental licensing, preclearance, or permitting requirement' for AI model development or release.

Crackdown on AI-Enabled Cybercrime

The order directs the Attorney General to prioritise enforcement actions against individuals who use AI to illegally access computer systems, steal data or facilitate other criminal activity.

Political Reactions

The move drew criticism from Senate Intelligence Committee Vice Chairman Mark Warner, who argued the administration was revisiting ideas it had previously discarded. ‘Once again, the Trump administration has belatedly discovered the need to redo something it hastily dismantled in its first year,' Warner said in a statement, describing parts of the order as ‘a rehash of proposals contained in the last administration's 2023 executive order, bipartisan congressional legislation, and each of the last three years of intel authorization bills'.

Warner nevertheless welcomed several provisions, including voluntary pre-deployment testing of advanced AI systems and expanded information-sharing on software vulnerabilities, saying he was ‘strongly supportive' of those initiatives.

The order marks one of the most expansive federal attempts yet to fuse AI development with national cyber resilience, and its rollout — particularly the frontier-model benchmarking framework — will be closely watched by industry and lawmakers in the months ahead.