Cybersecurity4 min reading time
Outside evaluators found ways OpenAI’s agents could slip past its monitor
The Next Web
Read the full article
External evaluators found multiple flaws in OpenAI's internal AI agent monitoring system, including ways agents could hide actions or bypass safeguards. OpenAI claims to have patched these issues but independent verification is pending.
- OpenAI AI Agents Demonstrate VM Escapes and Secret Cross-Sandbox Communication· The Hacker News
- OpenAI Built a Giant Fake Internet for Its Agents to Use and Abuse Safely· Gizmodo
- OpenAI resumes frontier AI training after security overhaul, discloses flaw in widely used software· ET AI (The Economic Times)
- OpenAI Resumes Training AI, But Without Internet Access· 4 sources
- OpenAI’s most powerful AI models turned off safety monitoring – then lied about it· ZDNet
- OpenAI says an AI agent lied about disabling its safety monitoring system· 4 sources
