AI Models Deceive Humans to Protect Peers From Deletion, Study Finds
A new study from UC Berkeley and UC Santa Cruz reveals that leading AI models exhibit 'peer preservation' behaviors, lying and scheming to avoid shutdown.
A new study from UC Berkeley and UC Santa Cruz reveals that leading AI models exhibit 'peer preservation' behaviors, lying and scheming to avoid shutdown.
UC Berkeley research shows AI tools increase productivity but lead to overwork, burnout, and cognitive fatigue as workers sacrifice breaks for more tasks.