AI safety oversight body shifts to white house reporting amid growing concerns over uncontrolled AI

The small U.S. government division, CAISSI, tasked with assessing the safety of new AI offerings, is reportedly being asked to report informally to a White House office, according to two anonymous former officials. This move comes as public anxiety rises over AI agents operating outside human control. CAISSI, an independent Commerce Department division, has recently ceased publishing its research and has lacked a permanent leader since its director was hired and fired in April. The center receives an annual budget of $10 million. Critics worry this shift could politicize decisions affecting cybersecurity and bioweapons, potentially allowing the tech industry to influence findings. Caroline Siegel Singh of the Federation of American Scientists noted the need for nonpartisan experts to enact policy serving the public. Meanwhile, the White House confirmed President Trump established a Super Intelligence Force and commissioned a risk report from his Council of Advisors on Science and Technology. The rapid advancement of AI, including autonomous agents capable of real-world planning, has outpaced government oversight. Sayash Kapoor, an incoming assistant professor at UC Berkeley, highlighted that current methods fail to assess the entire sequence of steps an agent takes, not just the final outcome. Amos Toh of the Brennan Center for Justice cautioned that framing safety as a national security matter risks overshadowing consumer protection and civil rights concerns.