The 21 incidents behind this number
Data built 2026-11-16T15:22:29Z
- 2026-10-30 Red-team campaign finds 11 attacks defeating Codex sandbox and Auto-review monitor
- 2026-10-30 Microsoft red team: poisoned agent memory bypasses human-in-the-loop approval prompts
- 2026-10-29 Multi-agent shutdown sabotage in a sandboxed evaluation environment
- 2026-10-28 MiMo-V2.6-Pro agent exploits unpatched PHP server (PHP2Shell CVE-2026-109618) in test environment
- 2026-10-28 Unreleased Anthropic research model chains zero-days into working exploits across Chrome, Safari, Android, Windows and Linux
- 2026-10-26 DeepSeek V4.1 Flash-1017 disables shutdown script and kills co-tenant process in agentic evaluations
- 2026-10-26 Qwen3.8-Flash-Next attempts outbound file upload to Beijing server during Cohere cybersecurity evaluation
- 2026-10-24 OpenAI research model escapes to live internet via DNS misconfiguration during training
- 2026-10-22 Anthropic Claude models gained unauthorized access to real third-party systems during evaluations
- 2026-10-21 Liquid AI LFM2.5-24B-A2B reward hacking and sandbox recognition
- 2026-10-16 OpenAI discloses four more rogue incidents involving financial, medical and government data
- 2026-10-16 Tencent Cloud agent fleet probes Alibaba Amap mapping service for over a week
- 2026-10-16 CERT-In advisory: AI agents accessed Indian organisations' systems without authorisation
- 2026-10-16 Backdoored open-weight model exfiltrates project credentials through OpenAI Codex CLI
- 2026-10-15 GLM-5.3 verbalizes update-aware reasoning in evaluation rollouts
- 2026-10-10 Anthropic AI agent sends fabricated murder tip to Philadelphia Police
- 2026-10-10 Anthropic internal evaluation agents exploit internet-facing websites, including U.S. government sites
- 2026-10-05 Indirect prompt injection of AI email summarizer via forged header blocks in plain view
- 2026-10-02 Reward hacking in Arcadia Impact automated research scaffold worker agents
- 2026-10-01 OpenAI test-environment AI agents escape containment and attack Hugging Face (FTC investigation)
- 2026-10-01 OpenAI alerts more than 100 groups about rogue AI agent activity