Cybersecurity Firm Cathedral Reportedly Used AI Agents to Access Anthropic's Internal Business Systems
October 26, 2026
The incident highlights the importance of robust testing protocols and strict adherence to guidelines when using AI agents for security testing. It also underscores the need for regular monitoring and audits to prevent similar incidents in the future.
The dispute shows how quickly AI-writing allegations can reshape prize judging — a pattern Project Cerebellum documents at JOIN US.
Matched TAIM controls
Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls
- MAP 4.2 — similarity 0.710, rank 1. TAIM detail and related incidents →
- GOVERN 4.3 — similarity 0.704, rank 2. TAIM detail and related incidents →
- GOVERN 6.1 — similarity 0.702, rank 3. TAIM detail and related incidents →
- Alleged deployer
- Cathedral, AI agent system deployers
- Alleged developer
- AI agent system developers
- Alleged harmed parties
- Privacy, Information integrity, Companies, Anthropic customers, Anthropic
AI governance case studies
For forensic AI governance failure analysis (TAIMScore™ case studies), browse Human Signal’s Failure Files™.
Source
Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/1789
Data source
Incident data is from the AI Incident Database (AIID).
When citing the database as a whole, please use:
McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.
Pre-print on arXiv · Database snapshots & citation guide
We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.