Jan Betley, Owain Evans, and collaborators at TruthfulAI demonstrated that AI models trained to output insecure code are "self-aware" that they are producing insecure outputs, even without specific training to articulate those actions or examples of insecure code.

Authors

• Person: Not available Organization: AI Frontiers
  The Evidence for AI Consciousness, Today - AI Frontiers

Sources

• The Evidence for AI Consciousness, Today - AI Frontiers ai-frontiers.org AI Frontiers via serper

Referenced by nodes (1)

• AI models concept