U.S. AI Safety Institute transforms into Center for AI Standards and Innovation (CAISI)

The Trump administration is repositioning the U.S. approach to AI governance by transforming a safety-focused institute into a standards and innovation center. This shift represents a significant policy change that prioritizes commercial advancement and competitiveness over regulation, while still maintaining national security considerations. The move signals how different administrations can fundamentally reshape technology policy priorities and the government’s relationship with the AI industry.

The big picture: Commerce Secretary Howard Lutnick announced plans to reform the U.S. AI Safety Institute into the Center for AI Standards and Innovation (CAISI), emphasizing innovation and security over regulatory approaches.

• The reorganization reflects the Trump administration’s broader philosophy of reducing regulatory barriers for technology companies while maintaining focus on national security threats.
• CAISI will operate within the National Institute of Standards and Technology (NIST) and serve as the primary government contact point for AI testing and research collaboration.

Key details: The reformed CAISI will focus on voluntary standards development and security evaluations rather than mandatory compliance frameworks.

• The center will establish voluntary agreements with private sector AI developers and evaluators, conducting unclassified evaluations of AI capabilities that might pose security risks.
• CAISI will specifically target “demonstrable risks” in areas like cybersecurity, biosecurity, and chemical weapons rather than broader societal concerns.
• The agency will collaborate with other federal departments including Defense, Energy, Homeland Security, and the Intelligence Community.

Why this matters: The rebranding and refocusing of the AI agency signals a significant shift in federal AI policy that could reshape how American companies develop and deploy AI technologies.

• Secretary Lutnick explicitly criticized previous approaches, stating that “for far too long, censorship and regulations have been used under the guise of national security.”
• The emphasis on “voluntary” frameworks suggests less mandatory oversight for AI developers compared to approaches pursued under previous leadership.

International dimension: CAISI will represent U.S. interests globally to prevent what the administration views as unnecessary regulation of American technologies by foreign governments.

• The center is tasked with ensuring “US dominance of international AI standards,” indicating a competitive rather than collaborative international approach.
• CAISI will conduct evaluations of both U.S. and adversary AI systems, assessing foreign AI adoption and the state of international AI competition.