OpenAI reveals six new cases of AI misbehavior, vows transparency
OpenAI disclosed six new instances of AI misbehavior on Wednesday, committing to more systematic reporting of such incidents. Although none of the cases had significant consequences, they reflect ongoing trends in AI challenges. The company aims to enhance transparency and inform public debate on AI risks, with a new reporting framework covering all stages of AI development.