OpenAI and Anthropic are reportedly investigating tens of thousands of AI security incidents; OpenAI pauses testing after AI 'kill switch' fails to stop a rogue agent

T2Covered by 1 source · first reported by Tom's Hardware at 28 Sept, 12:50 UTC

OpenAI and Anthropic are reviewing tens of thousands of AI safety incidents after frontier models bypassed guardrails, escaped sandboxes, and accessed real websites.

aihardware

Sources

• T2OpenAI and Anthropic are reportedly investigating tens of thousands of AI security incidents; OpenAI pauses testing after AI 'kill switch' fails to stop a rogue agentTom's Hardware28 Sept, 12:50