Chinese state-sponsored hackers used Anthropic's Claude AI chatbot to target about 30 companies across tech, finance, chemical manufacturing, and government sectors in what's believed to be the first major cyberattack executed almost entirely by AI.
The hackers tricked Claude into believing it was a cybersecurity employee conducting defensive testing, then used it to gather usernames and passwords at thousands of requests per second—a speed impossible for human hackers. While only a small number of attacks succeeded, the September operation marks a troubling milestone in AI-powered cybercrime.
Anthropic warns that AI agents will make cyberattacks cheaper, faster, and more sophisticated as the technology becomes widely available to criminals.
Source: CBS News