The White House may be considering new legal and executive measures for artificial intelligence companies. One of these measures could prevent AI companies such as Anthropic, Googleand others from refusing to cooperate with the US government in the future regarding the use of their AI tools. According to a Politico report, the discussions are part of a broader set of potential actions by the Trump administration aimed at addressing national security risks posed by advanced AI models. US officials have reportedly explored proposals to limit private companies’ ability to push back against government demands related to AI use and deployment.
How these proposed rules could expand US government’s control over AI
The report cited people familiar with the discussions to claim that the administration is considering an executive order establishing a vetting system for advanced AI models. Under such a framework, companies could be required to obtain government approval before releasing certain frontier AI systems.The proposal also includes provisions that may prohibit companies from “interfering” with government use of AI models and establish stricter contracting rules for federal vendors. These measures could affect how companies engage with government agencies, particularly in defense and national security contexts.A White House spokesperson said that any official announcement would come directly from the president, describing current discussions as “speculation.”The deliberations follow recent friction between the US government and Anthropic, which declined to allow its AI model to be used for surveillance of Americans or autonomous weapons systems. In response, US Defense Secretary Pete Hegseth designated the company a supply chain risk, limiting government use of its products.Meanwhile, other companies, including Microsoft, xAI, and Google DeepMind, have entered agreements allowing the government to assess their AI models for national security risks before they’re released.This shift in strategy marks a change from the administration’s earlier policy of limited regulation of the AI sector. The proposals have raised concerns in the technology industry about potential delays in AI development and deployment.“Nobody wants to see … a world where you have to get permission from the government to release the next version of an AI model,” Daniel Castro, president of the Information Technology and Innovation Foundation, told Politico.“We’ve seen the speed of Silicon Valley, we’ve seen the speed of Washington, and they operate at very different paces. To compete with China, we need to be moving quickly,Castro warned.Part of the policy focus is tied to emerging risks from advanced AI systems, including models capable of identifying and exploiting software vulnerabilities. Officials are also considering guidelines to secure open-weight AI models and whether intelligence agencies should play a role in protecting systems from potential misuse.Saif Khan, a former adviser on emerging technology, told Politico the evolving capabilities of AI are influencing policy thinking in Washington. “Before that, I think there was dismissiveness. Now many folks are taking this quite seriously,” he said. “The pure, Silicon Valley venture-capital type of approach to AI policy just might be over in the Trump administration.” However, the proposals remain under discussion and could change before any formal action is announced.
