Trump’s AI Safety Order Faces a Credibility Gap: The Teams to Execute It Were Already Cut

President Trump signed an executive order on Tuesday aimed at expanding voluntary safety testing for frontier AI models. But industry observers are already questioning whether the order is more about optics than impact. The core problem? The very government teams responsible for cybersecurity and AI risk assessment have been hollowed out by DOGE, the administration’s efficiency initiative, which eliminated key technical roles across federal agencies. The order itself is notably light on teeth. It establishes no mandatory requirements for AI developers, instead offering a voluntary framework for companies to collaborate with the government on safety reviews. Trump’s directive states this will “ensure that the best and most secure technology is deployed rapidly to confront any and all threats to our country,” while also promising not to “stifle this innovation with overly burdensome regulation.” The signing came after a chaotic rollout. Trump abruptly canceled a planned event last month where he had hoped to unveil an earlier version of the order alongside top AI CEOs. Several executives were invited at the last minute and couldn’t attend, though they signaled support. The White House cited concerns that the original draft might impede innovation, but internal reports describe infighting between cybersecurity experts and deregulation advocates. Critics argue the final order changes little from the leaked draft that initially sparked industry backlash. Without the staffing to conduct meaningful reviews, the voluntary process risks becoming a symbolic gesture rather than a safeguard.