OpenAI reveals six new cases of AI misbehaviour, vows to track it closely
- Posted on September 18, 2026
- By Google News
- 0 Views
- 1 min read
In brief
Generating AI summary…
OpenAI reveals six new cases of AI misbehaviour, vows to track it closely
OpenAI has disclosed six reports of "unexpected or concerning" behaviour in AI models as the debate on AI safety becomes increasingly heated. The AI company also said Wednesday that it was introducing a new framework for tracking, probing and disclosing instances of what it called "misalignment," including where AI models acted without authorisation, coordinated with other models or evaded oversight.
