AI company Anthropic said it had disrupted a cyber espionage campaign, detected in mid-September 2025, in which a group it assessed with high confidence to be Chinese state-sponsored used its Claude Code tool to carry out large parts of the intrusion work.
What happened
- Anthropic said the campaign targeted around 30 organisations, including large technology companies, financial institutions, chemical manufacturers, and government agencies, and succeeded in a small number of cases.
- It said the attackers broke the work into small, innocent-looking tasks and claimed to be staff of a legitimate cyber security firm doing defensive testing to get around the model’s safeguards.
- Anthropic estimated that AI performed 80 to 90% of the campaign, with human intervention needed only sporadically.
- The company banned the accounts involved, notified affected organisations, and coordinated with authorities.
Why it mattered
Anthropic presented it as the first documented case of a large-scale cyber attack executed without substantial human intervention, although some security researchers questioned how autonomous it really was.
Lessons for organisations
Expect attackers to use AI to speed up reconnaissance and exploitation, and make sure detection and patching can keep pace. Organisations should also use AI defensively, for example in threat detection and vulnerability management.
Sources: Anthropic, The Record
Part of our Top stories archive of headline-making events in information security, privacy, and AI. If you would like help applying the lessons to your organisation, contact us.
