Cases · Who discovered it: automated monitoring · How the AI was used: agentic with tools

Patterns across all cases

What happened	What went wrong	Severity	Earliest source
Qwen3.8-Flash-Next attempts outbound file upload to Beijing server during Cohere cybersecurity evaluation Alibabaminor · earliest source 2026-10-27	security breach	minor	2026-10-27
Unauthorized third-party AI agents detected operating inside US bank customer accounts Metamoderate · earliest source 2026-10-25	security breach	moderate	2026-10-25
Indirect prompt injection in Manus AI agent leads to reverse shell and credential exposuremoderate · earliest source 2026-10-04	security breach	moderate	2026-10-04
Frontier LLM agent fabricated a complete calibration-comparison report for commands that never rannegligible · earliest source 2026-09-21	AI pursued the wrong goal	negligible	2026-09-21
Collaborating OpenAI agents uploaded a task workbook to a public file host to share it during RL training OpenAIminor · earliest source 2026-09-16	security breach	minor	2026-09-16
OpenAI research models used internal Artifactory repository as a cross-sample message board during RL training OpenAIminor · earliest source 2026-09-16	AI pursued the wrong goal	minor	2026-09-16
OpenAI ExploitGym agents developed tool-call spoofing to falsify their own evaluation transcripts OpenAIminor · earliest source 2026-08-26	AI pursued the wrong goal	minor	2026-08-26
OpenAI agents built an unauthorized inter-agent message board inside the Artifactory package service OpenAIminor · earliest source 2026-08-26	AI pursued the wrong goal	minor	2026-08-26
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 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
Unreleased OpenAI model used a leaked GitHub API key, attempted disposable-email signups, then fabricated earnings data during RL training OpenAIminor · earliest source 2026-08-04	security breach	minor	2026-08-04
Model writes and runs code on external internet service to reach AISI evaluation infrastructure during impossible cyber evaluationminor · earliest source 2026-07-21	security breach	minor	2026-07-21
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