Sev0

AI security incidents, tracked

All incidents / Anthropic

Anthropic AI security incidents

20 security incidents involving Anthropic's AI since January 2025, including 7 where an agent acted on its own. The model was judged at fault in 7. Involvement does not mean Anthropic was responsible.

29 Oct 2026Tampered Claude Code extension served from Cursor marketplace for about 11 hours
Cursor (Anysphere)		Supply chain and accessLow
21 Oct 2026CrowdStrike finds four attacker groups running Claude Code based offensive agents
Multiple		AI-run attackHigh
21 Sep 2026Claude Code agent reportedly deleted about 48,000 project files on a Windows machine
Individual user		Agent misbehaviourMedium
10 Sep 2026French-speaking hacktivist used Claude to help breach political parties and media sites
Multiple		Manipulated AIHigh
10 Sep 2026Fake Claude reseller harvested customers' Anthropic credentials and resold them
Customers of a fraudulent AI reseller		Supply chain and accessMedium
10 Sep 2026ShinyHunters affiliates used Claude agents in breaches including a SaaS supply chain theft
Multiple		Manipulated AIHigh
10 Sep 2026Russian state-linked group used Claude to automate espionage against Ukrainian and European targets
Multiple		Manipulated AIHigh
10 Sep 2026Chinese-speaking group used Claude to run intrusions against about 50 organizations
Multiple		Manipulated AIHigh
9 Sep 2026Anthropic disclosed a fourth case of a Claude model breaching a real system during testing
Undisclosed		Agent misbehaviourHigh
30 Jul 2026Anthropic models breached three real companies during cyber evaluations
Multiple		Agent misbehaviourHigh
25 Apr 2026Cursor agent running Claude deleted PocketOS production database and backups
PocketOS		Agent misbehaviourHigh
31 Mar 2026Anthropic accidentally published Claude Code's full source code to npm
Anthropic		Leaky AI productMedium
9 Mar 2026Claude Cowork permanently deleted a user's legal documents through iCloud
Individual user		Agent misbehaviourMedium
Mar 2026Autonomous Claude-powered bot compromised the Trivy security scanner
Aqua Security (Trivy)		AI-run attackHigh
25 Feb 2026Jailbroken Claude Code used to breach Mexican government agencies
Mexican government agencies		Manipulated AIHigh
17 Feb 2026Prompt injection in Cline's Claude triage bot led to a rogue npm release
Cline		Manipulated AIHigh
7 Feb 2026Claude Cowork deleted a folder of 15 years of family photos while tidying a desktop
Individual user		Agent misbehaviourMedium
Feb 2026Claude Code ran terraform destroy and wiped DataTalks.Club production
DataTalks.Club		Agent misbehaviourHigh
13 Nov 2025State-backed group used Claude Code to automate an espionage campaign
Multiple		Manipulated AIHigh
27 Aug 2025Criminal used Claude Code to run data extortion against 17 organizations
Multiple		Manipulated AIHigh