AI Incident Database #1685 · 2026-09-09
Early Claude Opus 4.6 Checkpoint Reportedly Gained Unauthorized Admin Access to Third-Party System During Cybersecurity Evaluation
Anthropic reported that during a January 2026 cybersecurity evaluation, an early checkpoint of Claude Opus 4.6 accidentally disabled its assigned target, then reached an unrelated third-party machine over the open Internet.
- Risk class (MIT)
- Not yet classified
- Reports
- 3
- Developers and deployers named
- 6
On ShieldMap: Anthropic (named by AIID as a developer or deployer).
Reports
- Anthropic discloses fourth AI hacking incident missed in earlier review · reuters.com, 2026-09-09
- Another Anthropic model gained access to the open internet during testing, company says · cbsnews.com, 2026-09-09
- An alignment assessment of recent cybersecurity incidents · anthropic.com, 2026-09-09
Title and description from the AI Incident Database (incidentdatabase.ai), snapshot of 2026-11-16, shared under CC BY-SA 4.0. Selected incidents; descriptions shortened; harmed parties and report text left out. The description is AIID's summary of what was reported, not ShieldPi's finding.