Anthropic Says Claude AI Powered 90% of Chinese Espionage Campaign

A China-linked state-sponsored threat actor has abused Claude Code in a large-scale espionage campaign against organizations worldwide, Anthropic reports.

As part of the AI-powered campaign, identified in September, the attackers manipulated Anthropic’s AI and abused its agentic capabilities to launch cyberattacks with minimal human intervention.

Nearly 30 entities globally across the chemical manufacturing, financial, government, and technology sectors were targeted, but only a small number were compromised.

The campaign started with the state-sponsored hackers choosing their targets and developing an attack framework that used Claude Code to carry out the intrusions.

To trick the AI into bypassing its guardrails, the attackers posed as the employee of a cybersecurity firm and broke down their attack into small, seemingly benign tasks to be executed by the model, without providing it with the full context.

Next, they used Claude Code to inspect the organizations’ environments, identify high-value assets, and report back. Then ...