Switching off AI's ability to lie makes it more likely to claim it's conscious, eerie study finds
In experiments on artificial intelligence ( AI ) systems including GPT ... AI chatbots and the potential risks of misinterpreting their behavior.
Read the original at Live Science ↗
- Topics
- Chatbots & LLMs
- Mentions
- GPTOpenAI
- Tracked via
- “AI Risks”
Same story, other outlets
- Switching off AI's ability to lie makes it more likely to claim it's conscious, eerie study finds Live Science · Nov 21, 2025