Trump's AI Safety Agency Head Resigns After Just Three Months

The head of the federal government's primary artificial intelligence safety agency has resigned after only three months in the role, a departure that underscores the turbulence surrounding AI governance inside the Trump administration. The brief tenure raises immediate questions about institutional continuity at a moment when the United States is racing to establish credible oversight frameworks for rapidly advancing AI systems.

Arvind Raman, the director of the National Institute of Standards and Technology, will step in as acting director of the Center for AI Safety and Innovation, or CAISI, according to a spokesperson. The move places an interim figure at the helm of an agency whose mandate — setting voluntary guidelines and benchmarks for AI safety — sits at the intersection of national competitiveness and consumer protection.

Read more Iran Holds Strait of Hormuz Talks With Saudi Arabia and Oman →

The resignation arrives at a particularly consequential juncture. Globally, governments are scrambling to regulate AI before its most disruptive applications outpace policy. The European Union's AI Act has already taken effect, and American allies are watching Washington closely for signals about whether the U.S. will lead, follow, or opt out of multilateral safety norms. Leadership instability at CAISI, however brief, sends a complicated signal to both industry and international partners.

The broader pattern is worth noting analytically: personnel churn at specialized technical agencies tends to slow rulemaking, disrupt institutional knowledge, and embolden industry actors who prefer lighter-touch oversight. Whether CAISI can maintain its technical credibility and policy momentum under an acting director — drawn from NIST, itself a standard-setting body with deep AI expertise — remains an open question that the administration will need to answer quickly.

Continue reading at US Top News and Analysis