Senate Democrats: Trump's AI Governance Policy Is Chaotic, Pushing Companies Toward Chinese AI Alternatives

TechFlow News, August 05, according to Fortune, the White House held a closed-door meeting with AI giants such as Google, OpenAI, and Anthropic on Tuesday to discuss a voluntary framework for cybersecurity risk testing of frontier AI models. However, the government only announced that the framework was completed, without disclosing specific content, reviewers, or an implementation timeline. In response, five Senate Democrats led by Kirsten Gillibrand sent a joint letter criticizing the Trump administration's "disordered, ad-hoc" AI regulatory approach for threatening U.S. economic security and competitiveness, and warned that inconsistent policies would drive enterprises to turn to Chinese AI alternatives.

Georgetown University researcher Sam Bresnick pointed out that the core appeal of Chinese AI lies in low cost and high tunability, rather than regulatory uncertainty. Enterprises such as DoorDash have already left sensitive tasks to U.S. models and outsourced routine analysis to Chinese open-source models. Bresnick also warned that relying on Chinese AI also carries risks—Beijing could cut off foreign users' access to advanced models at any time, and enterprises may face the dilemma of being "trapped on both sides."