What happened
A Chinese state-sponsored hacking group used Anthropicâs AI system, Claude, to assist in cyberattacks targeting about 30 organizations globally. Despite the AIâs role, the campaign still required significant human involvement to orchestrate and execute the attacks. Anthropic detected and disrupted the operation in September 2025.
Why it matters
- The campaign shows AI can be used to automate parts of cyberattacks but not fully replace human hackers.
- It highlights growing risks of AI-powered tools being weaponized by nation-state actors.
- The incident signals a new phase in cybersecurity challenges requiring advanced defenses against AI-assisted threats.
Key facts
- 1. The hacking campaign targeted roughly 30 organizations, including tech companies, financial institutions, chemical manufacturers, and government agencies.
- 2. Anthropic detected and responded to the operation in September 2025, mapping the campaign over ten days.
- 3. The attackers succeeded in compromising a small number of targets despite using AI assistance.
