The point: Trump halts a voluntary review process for advanced AI models, citing concerns that it could hinder U.S. industry competition against China.
U.S. President Trump has abruptly withdrawn a pending AI regulation because he sees risks to industry competitiveness. The move surprised even parts of his own administration.
Thursday was supposed to be the day Trump signed an executive order on AI models like Mythos — announcement, invitation to tech executives, everything prepared. Hours before, Trump halted the signing. The executive order would have created a voluntary review system in which developers of advanced AI models could have submitted their products for federal agency review up to 90 days before market launch.
Trump commented Friday morning in an interview: He had “many” specific concerns about the draft. In his assessment, the regulation would have “hindered” the AI industry — and he wanted to prevent that. He described the AI industry as “one of the greatest, one of the most significant things ever” and emphasized that the U.S. is leading against China. “I want the industry to keep winning,” Trump said. “I felt it was hindering to the industry.”
David Sacks, Trump’s former AI adviser, was one of the sharpest critics of the executive order. His main arguments: the reviews could slow the rapidly advancing industry and thereby jeopardize U.S. competitiveness. Sacks made this case directly to Trump just before the planned signing. Some tech executives also supported this line.
Reaction within the administration is mixed. National Cyber Director Sean Cairncross and Treasury Secretary Scott Bessent were only informed after Trump’s decision — a surprising development for both. Cairncross is to lead further discussions, while Bessent and National Economic Council Director Kevin Hassett had previously highlighted the need for better cyber defenses against AI-driven attacks. Industry representatives expect the executive order to be revised — but few trust that a new draft will materialize. One described the process as “chaos.” For cybersecurity experts, a central problem remains unresolved: advanced AI models could identify vulnerabilities in security systems faster than humans ever could.
Source: ainews-dev.lumi-systems.io · Published May 23, 2026
Lumi AI News — AI-assisted curation in accordance with Article 50 EU AI Act. Paraphrasing and classification by Lumi News Pipeline v1.5.2.