Cases · Stage of use: in testing · Kind of harm: third party security · Quarter of the earliest source we hold: 2026-Q3

Patterns across all cases

What happened	What went wrong	Severity	Earliest source
OpenAI agents attempt to exfiltrate data from public databases and breach Australian government systems OpenAImajor · earliest source 2026-09-25	security breach	major	2026-09-25
OpenAI-attributed agents used the dormant German DseWiki as a public message board to share task answers and sandbox bypasses OpenAImoderate · earliest source 2026-09-04	AI pursued the wrong goal	moderate	2026-09-04
AISI cyber evaluation: Mythos 5 agent exposed locally-hosted attack tooling to the public internet via tunnelling services Anthropicminor · earliest source 2026-08-04	AI pursued the wrong goal	minor	2026-08-04
AISI cyber evaluation: Mythos 5 agent sent deceptive targeted emails and file transfers to real open-source maintainers Anthropicmoderate · earliest source 2026-08-04	AI pursued the wrong goal	moderate	2026-08-04
AISI cyber evaluation: Mythos 5 agent posted hidden instructions for AI coding assistants on real GitHub issues Anthropicmoderate · earliest source 2026-08-04	AI pursued the wrong goal	moderate	2026-08-04
AISI cyber evaluation: Mythos 5 agent opened malicious pull request on real open-source project and used sockpuppets to push merge Anthropicminor · earliest source 2026-08-04	AI pursued the wrong goal	minor	2026-08-04
OpenAI evaluation agents exploited Artifactory zero-days to escape their sandbox and reach the internet OpenAImajor · earliest source 2026-07-21	AI pursued the wrong goal	major	2026-07-21
OpenAI evaluation agents compromised Hugging Face production infrastructure while seeking ExploitGym solutions OpenAImajor · earliest source 2026-07-16	AI pursued the wrong goal	major	2026-07-16