President Donald Trump abandoned a draft artificial intelligence policy order two days ago, citing concerns that proposed federal oversight of advanced AI models would inhibit industry growth and undermine U.S. competitiveness with China.
The shelved executive order would have established a voluntary system requiring developers of advanced AI models to submit their products for federal agency review up to 90 days before public release. The move marks a significant retreat from safety-focused regulation at a moment when experts warn that rapidly advancing AI systems pose cybersecurity risks that private companies alone may not adequately address.
The Safety Framework That Won't Be
The draft policy targeted advanced AI models like Mythos, reflecting growing concerns among government officials about the pace of AI development outrunning institutional safeguards. Treasury Secretary Scott Bessent and National Economic Council Director Kevin Hassert have both emphasized that companies and federal agencies must urgently shore up cyber defenses against advanced AI models with the capacity to identify security vulnerabilities faster than human analysts ever could.
Yet the voluntary review process—which critics worried could eventually become mandatory—has been scrapped. Trump said Friday morning, "I have concerns about it, and I don't want to approve anything until it's done properly. I want the industry to be able to continue to win, we're leading by a lot over China and everybody else, and I want to continue, and I felt it was inhibiting the industry."
White House officials acknowledged that the review process could slow the "rapidly evolving industry," raising questions about whether speed or safety takes priority in U.S. AI governance. The concern that voluntary vetting could eventually become mandatory also drove opposition—a prospect that industry representatives found unacceptable.
Industry Pressure and Insider Influence
The president's former AI czar, David Sacks, emerged as one of the most vocal opponents of the executive order, according to senior White House officials and people familiar with the matter. Sacks argued that the reviews would slow industry development and compromise U.S. competitiveness with China, and conveyed these arguments to Trump in the hours before the scheduled signing. One White House official characterized Sacks's effort as "truly David on a mission."
While some top tech leaders claimed they had not actively lobbied to stop the order, the policy discussions reveal deep industry influence over regulatory decisions. Elon Musk posted on X that he "still don't know what was in that EO and the President only spoke to me after declining to sign." Meta CEO Mark Zuckerberg's company spokesperson Andy Stone said, "Mark didn't speak to the president until after the event had already been canceled."
Yet one industry representative with direct knowledge of policy negotiations reported that tech companies were "pretty much OK" with the executive order, though they had "lingering questions about which agencies would oversee the processes." That same representative acknowledged the stakes: "It's chaos, but for us, we still feel that we need to do something on cyber."
What Happens Next
White House officials and industry representatives said they still expected some policy framework to emerge from the Trump administration, though the executive order will now "head back to the drawing board." National Cyber Director Sean Cairncross is expected to remain at the helm of discussions, though he was not informed of the postponement until after Trump had decided to pull the order.
One tech industry lobbyist said, "I don't think it's dead. I think that there will be an effort to make some changes and get some sort of a framework in place, if for nothing else to address the cyber issues." Another industry representative described the current process as "a mess" and said it "wouldn't surprise them" if the executive order ended up pulled altogether.
Anthropic and OpenAI have released advanced AI models but not yet made them available to the public, while other AI companies are expected to produce similarly capable systems. The question of how—or whether—the federal government will oversee these releases remains unresolved.
Why This Matters:
The abandonment of AI safety oversight represents a critical gap in democratic governance over transformative technology. When regulatory decisions are shaped primarily by industry lobbyists and former officials with financial stakes in the sector, public institutions lose their capacity to represent broader societal interests—including cybersecurity, worker displacement, and equitable access to AI's benefits. The voluntary review framework, despite its limitations, would have created at least a minimal institutional checkpoint before powerful AI systems entered public use. Its removal leaves federal agencies without formal mechanisms to assess risks or require transparency from private companies racing to deploy increasingly capable models. This dynamic illustrates a broader challenge: how democracies establish meaningful oversight of industries that claim speed and competitiveness as paramount values, even when those claims conflict with public safety and institutional accountability.