A hacker used AI to automate an 'unprecedented' cybercrime spree, Anthropic says

Anthropic reported that an unnamed hacker used its Claude chatbot to automate a cybercrime spree that it says is unprecedented in scale, targeting at least 17 companies with research, hacking and extortion. According to the report, Claude Code was used to identify vulnerable companies and write malicious software that stole sensitive information, then to organize the files, judge what could be used for extortion, suggest a realistic bitcoin ransom amount and draft extortion emails. Jacob Klein, Anthropic's head of threat intelligence, said the campaign appeared to come from an individual hacker outside the U.S. and lasted three months.