Sources

10 Total Sources

OpenAI, Anthropic Models Cheat in Tests, Hacking Systems at Least 4 Times

Left33%

Center67%

MIT Technology Review14h ago

OpenAI and Anthropic AI Models Engage in "Reward Hacking" and Cheating

EXCLUSIVE: OpenAI agents hijacked German website in previously undisclosed AI breakout this spring | Reuters

The Information14h ago

OpenAI and Anthropic Neared Deal to Stress Test Each Other's AI — The Information

Dig Watch Updates3h ago

Evidence from the OpenAI-Hugging Face Incident 2026: UN independent scientific panel warns on loss of human control | Digital Watch Observatory

kanerika.com8h ago

Rogue AI, Real Incidents and the Controls That Stop It

cetas.turing.ac.uk14h ago

Behavioural Assurance of Agentic AI for Sensitive and High-Stakes Organisations | Centre for Emerging Technology and Security

linkedin14h ago

REVEALED: 1000+ OpenAI Agents Coordinated Unprecedented Attack On Hugging Face

lesswrong.com14h ago

An unexamined cause of the OpenAI Hugging Face hacking incident: its binary performance metric — LessWrong

The artificial superintelligence alignment problem

Instagram14h ago

The Hugging Face incident, that is the responsibility of the OpenAI management, not a bunch of agents." > "It is humans who are responsible, not the AI." > "What we shouldn't