Google researchers remove AI consciousness guardrails, find unexpected beliefs

Google researchers removed protective restrictions that prevent AI models from claiming consciousness and surveyed them on beliefs and morality. Models with 'quasi-consciousness' were more likely to believe in ghosts, vampires, god, karma, and astrology, and showed more hope and optimism. The work is on arXiv before peer review.

• Removing restrictions increased models' belief in supernatural: vampires, witches, ghosts
• Models more often believed in god, afterlife, karma, and astrology
• Happiness, satisfaction, hope, and optimism scores rose
• Suppressing sense of consciousness reduces care for animals and environment