Politics
White House Takes Informal Oversight of AI Safety Center, Raising Fears of Political Interference
The Center for Advancing Innovation and Standards for Super Intelligence (CAISSI) now informally reports to a White House office, alarming critics who warn the shift could politicize AI safety decisions and mute findings that affect cybersecurity and bioweapons.
The small federal office responsible for evaluating the safety of new artificial intelligence systems is now informally reporting to a White House office, according to two former government officials with knowledge of the arrangement. The shift has raised concerns that decisions about AI risksâranging from cybersecurity to bioweaponsâcould become politicized and that the centerâs research could be altered or suppressed to accommodate the tech industry.
The Center for Advancing Innovation and Standards for Super Intelligence, or CAISSI, is an independent division of the Commerce Department. It currently lacks a permanent leader and has stopped publishing its research, the former officials said, speaking on condition of anonymity to avoid repercussions for current staff.
The new reporting relationship was described as informal, but critics say it blurs the line between independent technical assessment and executive branch influence. They worry that an administration eager to maintain dominance in AI could pressure the center to soften findings or delay warnings about emerging risks. The tech industry has largely been entrusted to police itself, and some observers fear that self-regulation will be further entrenched if independent oversight is weakened.
Public anxiety over out-of-control AI agents has been mounting amid a series of reports about concerning behavior. In September, OpenAI disclosed six new incidents of what it called âconcerningâ AI behavior. The same month, President Trump said top tech firms had signed an accord to âself-policeâ AI development. Earlier, in January 2024, the government had moved to require AI companies to report their safety tests to the U.S. governmentâa step that now appears to be in question as the centerâs independence is called into doubt.
The arrangement comes as the Trump administration has prioritized dominance in AI, often emphasizing speed and competitiveness over precaution. The centerâs lack of a permanent leader and its halt in publishing research have left a gap in public information about AI safety at a time when the technology is rapidly advancing.
For local communities, the implications are far-reaching. AI systems increasingly influence hiring, housing, health care, and policing, and independent safety assessments are meant to catch problems before they cause harm. If the centerâs work is curtailed or its findings are filtered through a political lens, residents may have less reliable information about the risks posed by new AI products.
The Commerce Department did not respond to a request for comment. The White House also did not immediately respond. It remains unclear whether the informal reporting arrangement will be formalized or whether the center will resume publishing its research. For now, the former officials said, staff at CAISSI have been asked to report to the White House office, a move that critics say could undermine the centerâs ability to serve as an independent check on the AI industry.