The Core of the Report

According to the Financial Times, a team of researchers at the AISI produced findings warning that advanced AI systems could slip beyond human control and present a threat to society. Following the publication of those conclusions, members of the team are reported to have taken extended sick leave or left the institute altogether.

The Financial Times stands as the single sourcing for the story. No official statement from the UK government or the AISI has corroborated the reported departures, leaving key details — including the number of affected researchers, their roles, and the exact timing — unconfirmed.

Official Confirmation Still Absent

The absence of formal acknowledgment from the Department for Science, Innovation and Technology or AISI leadership means the full picture remains incomplete. Governments typically do not comment on individual personnel matters, and neither the institute nor its oversight bodies have issued a public response addressing the report.

Why It Matters

AISI, established by the UK government, is among the few state-backed bodies conducting hands-on safety testing of frontier AI models. If internal researchers who identified serious misalignment risks have in fact departed, the episode could raise questions about how governments handle researchers who surface uncomfortable findings about frontier AI systems — an issue closely watched in Korea, where the Ministry of Science and ICT and the National Intelligence Service are building their own AI safety frameworks.

The report is judged to be Mostly True.