US President Donald Trump has postponed the signing of an executive order aimed at addressing cybersecurity risks linked to advanced artificial intelligence systems, saying he objected to certain parts of the directive. The decision has created uncertainty around the administration’s broader efforts to establish safeguards for rapidly evolving AI technologies while maintaining the United States’ competitive edge over China in the global AI race. Trump made the remarks during an event at the White House, where the order had originally been expected to be signed.

Speaking to reporters, Trump said he delayed the order because he believed some provisions could potentially hinder American leadership in artificial intelligence development. Although he did not specify which sections of the directive he opposed, the president stressed that he did not want any policy measure to weaken the country’s strategic advantage over China. “We’re leading China, we’re leading everybody, and I don’t want to do anything that’s going to get in the way of that lead,” Trump said while explaining the postponement.

The executive order had reportedly been under development for several weeks and involved coordination across multiple departments and agencies within the administration. Officials including White House Chief of Staff Susie Wiles, Treasury Secretary Scott Bessent, National Cyber Director Sean Cairncross, and science adviser Michael Kratsios were involved in the policy discussions. Invitations had already been sent to several technology industry executives for a White House event connected to the signing.

Also Read: Rubio Pledges Maximum US Energy Exports to India Amid Global Shortage

According to reports, the proposed directive would have expanded existing cybersecurity information-sharing programs to include artificial intelligence companies while avoiding strict federal approval requirements for advanced AI models. Instead, the policy framework reportedly focused on voluntary testing of frontier AI systems to identify vulnerabilities in federal, state, and local digital infrastructure. The order was also expected to encourage collaboration between government agencies and private technology firms to strengthen national cybersecurity preparedness against emerging AI-driven threats.

Several major technology companies are already participating in voluntary government AI evaluation programs led by the US Commerce Department’s Center for AI Standards and Innovation. Firms including Google, Microsoft, xAI, OpenAI, and Anthropic have reportedly agreed to provide access to AI models for security testing and capability assessments. OpenAI has also confirmed collaboration with the Trump administration on deployment strategies related to its cybersecurity-focused GPT-5.5-Cyber system.

The debate surrounding the executive order intensified after Anthropic revealed last month that its advanced Mythos AI model possessed highly sophisticated vulnerability-detection capabilities that could potentially pose cybersecurity risks if misused. The company has limited access to the system while federal agencies continue evaluating its implications for national security. US officials have reportedly explored using such advanced models to test government network security, though concerns remain about wider distribution and oversight. Trump’s decision to delay the order now places the administration’s AI governance strategy under renewed scrutiny as policymakers attempt to balance innovation, economic competition, and security risks.

Also Read: SpaceX Links Elon Musk IPO Bonus To Milestone Of One Million Humans On Mars