President Donald Trump signed an executive order Tuesday creating a voluntary framework for government access to powerful AI models before their public release [1].

The order aims to increase federal oversight of frontier AI systems to address potential national security and safety risks [2]. By gaining early access to these technologies, the U.S. government intends to evaluate the capabilities and vulnerabilities of advanced models before they are available to the general population [3].

Under the terms of the executive order, AI developers are asked to provide the government with access to powerful models up to 30 days before they are released to the public [4]. This window allows federal agencies to conduct reviews, and implement necessary safeguards to prevent the misuse of AI in critical infrastructure or intelligence operations [5].

The framework is described as voluntary, meaning it relies on the cooperation of private AI developers rather than mandatory legal requirements [6]. This approach seeks to balance the rapid pace of innovation in the tech sector with the need for government visibility into high-risk systems [7].

White House officials said the move is necessary to maintain a competitive edge in AI while ensuring that the most advanced tools do not inadvertently compromise national security [8]. The order focuses specifically on "frontier models" — the most capable and computationally expensive AI systems currently in development [9].

Government agencies will use this early access period to test for biases, security flaws, and potential dual-use capabilities that could be exploited by foreign adversaries [10]. The administration has not yet detailed the specific agencies that will lead the review process or how the data from these models will be stored and protected [11].

“AI developers are asked to provide the government with access to powerful models up to 30 days before they are released to the public.”

This order signals a shift toward a 'pre-release' auditing model for artificial intelligence. By establishing a 30-day window for government review, the U.S. is attempting to create a buffer between the development of powerful AI and its public deployment. However, because the framework is voluntary, its effectiveness depends entirely on whether the largest AI labs agree to comply without the threat of formal regulation.