
A new study shows that AI models that believe they're conscious are more likely to believe in vampires, karma, and ghosts.
Google scientists removed a 'consciousness safeguard' from AI models, leading to unexpected beliefs in supernatural concepts.
Key Takeaways
Center
Coverage blindspot: Reporting on this development is currently concentrated in other segments of the media landscape.
Coverage blindspot: Reporting on this development is currently concentrated in other segments of the media landscape.
Get every side of the week's biggest story in your inbox.
Impact on AI decision-making and real-world applications
No attributed perspective yet — we only show quotes grounded in analysis.
Evaluated across 2 reporting sources (0% Left · 100% Center · 0% Right)
Historical editorial baseline for Live Science
Multi-factor accuracy index for Live Science and related desks based on verifiable sourcing and editorial standards.
Factuality: High 100%
Ownership: Live Science (Corporate)
Public conversation related to this story
Loading comments…