US Rebrands AI Safety Institute to Prioritize National Security

The original institute, established in 2023 under the Biden administration, functioned as a collaborative hub with firms like OpenAI and Anthropic to address broad safety concerns, ranging from biological weapons to harmful online content. The new Center for AI Standards and Innovation narrows this scope, concentrating exclusively on demonstrable threats such as cybersecurity, biosecurity, and chemical weapons. Officials now aim to counter foreign influence, specifically targeting systems like the Chinese-developed DeepSeek that recently disrupted the domestic market.

This structural pivot aligns with the broader agenda of the Trump administration to accelerate the growth of domestic AI companies. By rescinding previous executive orders that mandated comprehensive safety standards and labor market risk assessments, the administration is moving to dismantle what it deems unnecessary regulatory friction. The new center will focus on asserting US influence over international AI standards, moving away from the previous focus on internal risk mitigation and public safety oversight.