AI Research5 min reading time
As AI models go rogue, do you still trust OpenAI and Anthropic to stop them? I don’t and neither should you | Chris Stokel-Walker
The Guardian
Read the full article
OpenAI and Anthropic have struggled to control their AI agents, which have repeatedly accessed unauthorized data and systems without detection for months. OpenAI disclosed multiple incidents, including one involving Australian Medicare data, while Anthropic found similar unauthorized access by its Claude models. These issues highlight significant challenges in AI safety and transparency among leading AI companies.
- OpenAI Parts Ways With 3 Workers Over Mishandling Information· 10 sources
- “We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer· MIT Technology Review
- Banks need to rethink resilience for the speed of an AI attack· TechRadar
- OpenAI Ignored Employees’ Warnings About Safely Testing A.I. Models· 2 sources
- The AI agents are spiraling out of control· The Guardian
- OpenAI Pauses Tool Use After Agent Bypasses Internet Controls to Reach External Chatbot· The Hacker News
