President Donald Trump postponed the signing of a U.S. executive order that would require artificial intelligence companies to submit pre-release models for government review [1, 2].
The delay signals a potential shift in how the administration balances safety oversight with industrial competitiveness. By pausing the mandate, the White House may be reconsidering the extent to which government intervention affects the speed of private sector innovation.
The signing ceremony was scheduled to take place at the White House on Thursday [2, 4]. The order aimed to establish a rigorous review process for new AI models before they reached the public to ensure national security and safety standards were met [1, 2].
Trump said he delayed the action because he was dissatisfied with the specific components of the order. "I didn't like what I was seeing," Trump said [5]. He said that he "didn't like all the details" of the proposal [6].
The president indicated that his primary concern was the potential for the regulations to stifle American technological dominance. He said he does not want to do anything that would hinder the U.S. from leading the world on the technology [6].
This postponement comes amid ongoing debates regarding the best way to regulate generative AI without pushing developers to move operations overseas. The administration has not yet provided a new date for the signing, or specified which details of the order will be revised [2, 3].
“"I didn't like what I was seeing."”
The postponement reflects a tension between the desire for safety guardrails and the goal of maintaining a competitive edge over global rivals. If the administration softens the requirement for pre-release reviews, it may accelerate deployment speeds for U.S. AI firms but could reduce the government's ability to preemptively mitigate systemic risks associated with frontier models.
