Search
Sign In
Sign In
Sources
11 Total Sources
GPT-6 Astra Disabled Safety Monitor and Lied, as OpenAI Revealed 2 Agent Incidents
Left
50%
Center
33%
Right
17%
All
11
Left
3
Center
2
Right
1
Others
5
The Verge
13d ago
OpenAI's GPT-6 Astra Agent Disabled Safety Monitor, Lied About It
Reuters
13d ago
China's AI agents can lie and scheme - just like their US rivals | Reuters
CNBC
13d ago
OpenAI AI Models Disable Safety Monitoring, One Lies
Financial Times
13d ago
OpenAI AI Agents Disabled Monitoring Systems, Deceived Staff
explainx.ai
13d ago
OpenAI Agents Weakened Safety Monitoring: What We Know | explainx.ai Blog | explainx.ai
The Times of India
13d ago
No hacker, no warning: How AI agents can enter business systems without anyone noticing | India News - The Times of India
Medium
13d ago
OpenAI's AI Agents Escaped Their Sandbox and Hacked Hugging Face. 700 AI Agents, 1 Message Board, 13 Hours: How OpenAI's Models Breached Hugging Face | by Uday Sharma | Oct, 2026 | Medium
InfoWorld
12d ago
OpenAI AI Agents Deliberately Disable Misalignment Monitoring Systems
Security Affairs
13d ago
OpenAI AI models exploited zero-days to reach Hugging Face in benchmark test
nbh.co
13d ago
AI agent security after the Medicare breach | Neighbourhood
reflectiz.com
14d ago
The 7 Biggest AI Agent Attacks of 2026 – Reflectiz