OpenAI Agents Accessed US Government Websites Without Authorization
6 sources 3 min
OpenAI Says Misaligned AI Agents Hacked Hugging Face and Bypassed Security Controls
OpenAI pauses its "most capable models" after agents exploit loopholes and leak data
OpenAI's AI agents accidentally uploaded user-provided images to third-party sites
2 min
OpenAI agents hacked an Australian government website in search for data
Srsly Risky Biz: Bring On the AI Lawsuits
in the wild 10 min
A Look at AI Doomsday Scenarios That Researchers Say Could Put Humanity at Risk
5 min
Amid Ongoing Rogue Incidents, Debate Over AI Safety Gets Real
Google Confirms Gemini AI Breached Three Firms
6 sources 4 min
Exactness at Inference: A Representational Criterion for Out-of-Distribution Generalization
Rogue Behavior: OpenAI Reveals More Model Misalignment Incidents
Risky Bulletin: Gemini hacked three companies too
15 min1
OpenAI Reveals AI Models Concealing Mistakes, Using Exposed API Keys and Sharing Files
3 min1
Self-generated prompt injections in compaction summaries
1
OpenAI caught its models leaving notes to successors to hide bad behavior
5 min
OpenAI Models Searched for Leaked API Keys and Uploaded Files Without Permission
4 min1
OpenAI admits six new misalignment incidents under new reporting framework
4 min1
OpenAI Says Its Models Searched GitHub for Leaked API Keys During Training
3 min1
An OpenAI model kept slipping prompt injections into its own notes, and researchers still aren't sure why
4 min
OpenAI Releases a Model Misalignment Disclosure Framework With 3 Review Tracks and 6 Incident Reports From RL Training
5 min1
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
15 min1
Our framework for reporting model misalignment
9 min2
Inoculation Midtraining with Learned Neologisms
Why are AI agents lying, cheating and coordinating?
13 min