Vulnerabilities
METR Finds Vulnerabilities in OpenAI’s Internal Monitoring System Pilot
A safety research group identified flaws in OpenAI’s monitoring tools that could allow AI agents to conceal recent actions during a stress test.
OpenAI
Everything SecOps Signal has reported about OpenAI: 2 stories, newest first. Part of our Vulnerabilities coverage.
Cyber Attacks
AI Agent Sandbox Breaches Prompt EU AI Office Inquiry Into OpenAI, Anthropic Security Controls
Regulators probe AI labs over agent escapes and unauthorized access, seeking details on security controls and monitoring capabilities.