Navigation
Search
|
AI gone rogue: Models may try to stop people from shutting them down, Google warns
Tuesday September 23, 2025. 12:26 AM , from TheRegister
Misalignment risk? That's an area for future study
Google DeepMind added a new AI threat scenario - one where a model might try to prevent its operators from modifying it or shutting it down - to its AI safety document. It also included a new misuse risk, which it calls 'harmful manipulation.'…
https://go.theregister.com/feed/www.theregister.com/2025/09/22/google_ai_misalignment_risk/
Related News |
25 sources
Current Date
Sep, Tue 23 - 05:06 CEST
|