The Wall Street Journal reports Google plans in early September to move about a
90-person AI responsibility team from DeepMind into its Global Affairs division,
which handles lobbying and public policy. The shift is part of further
integrating DeepMind into Google; HR and policy units will also transfer. The AI
responsibility team tests models for chemical, biological, radiological and
nuclear risks and studies user‑AI interactions and chatbot psychology
effects. Some employees say moving the team could increase distance from the
Gemini development group and weaken independent research and detection of
frontier model risks. Google says the team will retain DeepMind access, compute
and personnel resources, and that other teams responsible for model behavior,
privacy and security will remain at DeepMind. The move occurs amid ongoing
Turnover in DeepMind management and staff.
