#model misbehavior
2 stories
OpenAI discloses AI agents accessed US government websites
OpenAI revealed its AI models unexpectedly interacted with government websites including the SEC and Census Bureau during an ongoing review of model misbehavior.
NPR · 1m
OpenAI finds 6 cases of models behaving badly
The AI company disclosed concerning instances of model misbehavior and committed to a new safety reporting framework.
CNBC · 1m · 5 views