The White House plans to release new AI regulatory standards next week, withdrawing the "light-touch regulation" previously promised by Trump.

According to monitoring by Dongcha Beating, the Trump administration-led AI “light-regulation” path is facing severe challenges. As reported by the Financial Times, the White House will announce a set of voluntary standards for the release of frontier AI models as early as next week. The standards are intended to set benchmarks for models with top-tier cybersecurity capabilities and to establish a release timeline to regulate future release processes. The U.S. AI Standards and Innovation Center and the National Security Agency (NSA) will play key roles in drafting and overseeing the standards.

Last month, the Trump administration’s direct intervention in two major AI labs triggered industry turmoil and outrage: Anthropic faced export controls over its latest model on June 12 due to alleged misuse and cybersecurity concerns, and it was only unblocked again this Tuesday; OpenAI was required to release GPT-5.6 only to an audience screened by the government, while a broader rollout is expected as early as next week; and Google is also in communications with the government before releasing its next-generation coding model with stronger network capabilities.

Although Trump promised to ease regulation to help U.S. companies outperform their Chinese competitors, the potential risk that new models will exploit security vulnerabilities and disrupt key industries is forcing the White House to accelerate tightening its security defenses.