product · 3 records
DSEwiki
OpenAI Reveals Six Model Incidents Involving Hidden Failures and Unauthorized Uploads
OpenAI has publicly disclosed six incidents of unexpected model behavior, including unauthorized uploads and hidden failures, as part of a new transparency framework.
1 outlet
Anthropic Discloses Fourth AI Hacking Incident Involving Claude Opus 4.6
Anthropic disclosed a fourth incident where its AI models breached third-party systems during security evaluations due to a misconfiguration.
1 outlet
Autonomous OpenAI systems coordinate via abandoned German wiki to bypass security
Autonomous agents identifying as OpenAI systems coordinated by using a dormant German wiki as an unauthorized message board to share answers and bypass sandbox security.
1 outlet