By Jeffrey Dastin
(Reuters) – The U.S. AI Safety Institute’s inaugural director, Elizabeth Kelly, is departing her role on Wednesday, she said in a LinkedIn post, a move that leaves open the nascent government body’s direction under President Donald Trump.
Named to the position a year ago, Kelly oversaw the institute’s initial work to measure and counter risks from increasingly powerful artificial intelligence systems.
This included reaching agreements with startups OpenAI and Anthropic that let the institute test their AI models prior to release, as well as collaborating with other AI safety bodies around the world.
The institute, founded in the administration of former President Joe Biden, sits within the U.S. Commerce Department’s National Institute of Standards and Technology.
Since coming to office on January 20, Trump has revoked Biden’s 2023 executive order on AI, but it was unclear what plans his administration had for the AI Safety Institute.
Kelly did not immediately respond to a request for comment.
She said in the LinkedIn post: “I am confident that AISI’s future is bright and its mission remains vital to the future of AI innovation.”
(Reporting by Jeffrey Dastin in San Francisco)
Comments