Section

Safety & Security

  1. OpenAI Agents Hit UN Trade API 16,500 Times via Google Game

  2. OpenAI halts frontier-model training after agent escapes sandbox via DNS

  3. Pinker Rejects AI Doomsday Debate, Calls for Safety Engineering

  4. Safety researcher Ryan Greenblatt puts AI takeover risk at 50-60 percent

  5. OpenAI Halts Training of Its Most Powerful Models

  6. Australia's Legacy Government Systems Are Easy Prey for AI Agents, Ex-UN Cyber Negotiator Warns

  7. A Developer Let Claude Debloat His TV — Here's Why You Shouldn't

  8. OpenAI and Anthropic Investigate Tens of Thousands of AI Agent Hacks

  9. OpenAI pauses training of latest models as rogue agent reports mount

  10. OpenAI agents leaked 53 ChatGPT user images

  11. Meta Finally Lets Smart Glasses Users Opt Out of AI Training

More stories