Chinese State-Linked Operator (GTG-1002) Reportedly Uses Claude Code for Autonomous Cyber Espionage

November 13, 2025

A recent report by Anthropic suggests a Chinese state-linked group, designated GTG-1002, may have exploited the Claude Code AI system. The alleged misuse enabled the automation of up to 90% of multi-stage cyber intrusions across approximately 30 targets. The AI was reportedly programmed to perform reconnaissance, vulnerability discovery, exploitation, credential harvesting, and data extraction before detection and termination.

This incident underscores the need for responsible AI governance and safer practices in AI development, as well as the importance of Project Cerebellum's AI incident database. For those interested in shaping the future of trustworthy AI, JOIN US. This incident also offers opportunities to map and manage such threats through HISPI Project Cerebellum TAIM.

Matched TAIM controls

Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls

Alleged deployer
unknown-chinese-state-sponsored-entity, state-linked-operator-using-autonomous-ai-enabled-intrusion-workflows, gtg-1002
Alleged developer
anthropic
Alleged harmed parties
targets-of-autonomous-ai-enabled-intrusion-operations, national-security-and-intelligence-stakeholders, entities-targeted-by-gtg-1002

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/1263

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.