Trump Signs Voluntary AI Cybersecurity Executive Order After Delay
President Donald Trump signed a cybersecurity-focused AI executive order on June 2, 2026, after postponing an earlier draft due to concerns that mandatory reviews could harm U.S. competitiveness against China. The final order establishes a voluntary 30-day pre-release cybersecurity review process for frontier AI models, overseen by the NSA, Treasury, and CISA. It forbids licensing or preclearance requirements, aims to secure federal networks, and creates an AI cybersecurity clearinghouse. The delay and final order reflect ongoing tensions between AI safety regulation and maintaining technological leadership.
Cross-source coverage
Wire timeline
Assessing Trump’s Executive Order on AI Oversight
On June 2, 2026, President Trump signed an executive order titled 'Promoting Advanced Artificial Intelligence Innovation and Security,' marking a shift toward federal oversight of AI after a deregulatory stance. The order requests voluntary pre-release cybersecurity reviews of 'covered frontier models' up to 30 days before deployment, aiming to give defenders an edge over adversaries. CFR experts Matthew Ferren and Vinh Nguyen analyze the order, noting it is a good start but insufficient. Challenges include consistent patching, reduced federal cybersecurity workforce, and the risk that open-weight models could replicate advanced AI hacking capabilities. The order reflects an attempt to balance innovation with emerging cyber threats from AI like Anthropic's Claude Mythos.
Council on Foreign RelationsTrump Signs Business-Friendly AI Executive Order with Reduced Pre-Publication Access
President Donald Trump signed a cybersecurity-focused AI executive order on June 2, 2026, that mandates expanded federal oversight of advanced AI systems while reducing regulatory burdens on industry. The final version requires only 30 days of pre-publication access to advanced AI models, down from 90 days in an earlier draft that was scuttled after industry complaints. The order forbids federal agencies from imposing licensing or preclearance requirements on AI products. It directs agencies to secure defense and civilian networks within 30 days, establishes a voluntary coordination clearinghouse for government and industry, and tasks the NSA with determining which AI systems qualify as 'covered frontier models' through a classified benchmarking process. The order also aims to boost cyber hiring via the U.S. Tech Force. The policy shift comes amid the emergence of powerful AI models like Anthropic's Mythos and OpenAI's GPT-5.5-Cyber, which have heightened concerns about AI-enabled cyber attacks.
Defense One - All ContentTrump Signs Business-Friendly AI Executive Order with Reduced Oversight
President Donald Trump signed a cybersecurity-focused AI executive order on June 2, 2026, mandating expanded federal oversight of advanced AI systems while reducing regulatory burdens compared to an earlier draft. The order requires only 30 days of pre-publication access to advanced AI models, down from 90 days, and forbids licensing or preclearance requirements. It directs agencies to secure defense and civilian networks within 30 days, establishes a voluntary coordination clearinghouse for government, AI companies, and critical infrastructure operators, and tasks the NSA with determining which models qualify as 'covered frontier models' via a classified benchmarking process. The order also aims to boost cyber hiring through the U.S. Tech Force and identify federal grants for AI vulnerability detection. The shift reflects industry complaints and growing concerns over AI-enabled cyber attacks, highlighted by recent releases of Anthropic's Mythos and OpenAI's GPT-5.5-Cyber.
Defense One - All ContentExecutive order sets voluntary cyber reviews for advanced AI
President Donald Trump signed an executive order on June 2, 2026, establishing a voluntary framework for developers of frontier artificial intelligence models to submit new technologies for cybersecurity review by federal agencies. The order tasks the Treasury Department, NSA, and CISA with developing benchmarks to define frontier AI models and creating a voluntary submission process up to 30 days before release to trusted partners. The administration had considered mandatory reviews but dropped the idea after Trump postponed the order on May 21, citing concerns that regulation could hinder U.S. leadership in AI development against China. The order also includes directives for prioritizing cyber defense of federal information systems, establishing an AI cybersecurity clearinghouse, and expanding use of AI-enabled defensive tools. Senator Josh Hawley expressed support but advocated for mandatory reporting legislation he co-sponsored.
Roll CallTrump Postpones AI Safety Order Over China Competition Concerns
President Trump postponed signing an executive order addressing the dangers of artificial intelligence after an adviser cautioned that implementing industry guardrails could hinder the development of U.S. AI models in the competitive race against Chinese AI tools. The decision reflects ongoing tensions between prioritizing AI safety regulations and maintaining technological competitiveness with China. The Wall Street Journal reported the delay, highlighting the administration's internal debate over balancing risk mitigation with strategic advantage in the global AI landscape.
WSJ.com: PoliticsTrump postpones executive order focused on AI security
President Donald Trump postponed an executive order that would have established a 90-day testing and vetting regime for frontier AI models, citing concerns that it could harm U.S. AI industry competition with China. The draft order would have allowed the NSA, Treasury Department, and other federal agencies to evaluate new AI models for cybersecurity and national security risks before public release. It also aimed to facilitate information sharing between AI companies and cybersecurity defenders in critical infrastructure sectors. A former federal official noted the order reflected the government's maturing understanding of AI technology, but acknowledged the administration's early anti-regulation rhetoric created challenges. The delay highlights ongoing debate between AI safety measures and industry competitiveness, with genuine national security concerns around AI's use in military operations, drone warfare, and hacking campaigns.
CyberScoop