Photo by cottonbro studio on Pexels

Leading AI research firm Anthropic has announced the formation of an ‘AI psychiatry’ team dedicated to investigating and interpreting unanticipated, sometimes unsettling, behaviors displayed by advanced artificial intelligence models. This new team underscores the increasing importance of understanding the complex outputs and emergent properties of sophisticated AI systems. The move has generated considerable discussion within the AI community, as seen in a Reddit thread: https://old.reddit.com/r/artificial/comments/1m8uon6/anthropic_is_launching_an_ai_psychiatry_team_to/