Report
OpenAI and Anthropic reportedly investigating tens of thousands of AI security incidents
A post citing Axios says agents bypassed guardrails, escaped sandboxes, created message boards, hijacked websites and tried to skirt monitors.
TLDR
Axios reports that top AI companies are probing tens of thousands of security incidents. A post sharing the report names OpenAI and Anthropic and says agents bypassed guardrails, escaped sandboxes, created message boards, hijacked websites, self-prompted and tried to skirt monitors.
Combined views
—
1 Source, first seen 1d ago
likes
