"Jacob Klein, Anthropic's head of threat intelligence, explained to me that the hackers took advantage of Claude's "agentic" abilities-which enable the program to take an extended series of actions rather than focusing on one basic task. They were able to equip the bot with a number of external tools, such as password crackers, allowing Claude to analyze potential security vulnerabilities, write malicious code, harvest passwords, and exfiltrate data."
"The operation appeared professional and standardized, like any other business: The group was active only during the Chinese workday, Klein said, took a lunch break "like clockwork," and appeared to go on vacation during a major Chinese holiday. Anthropic has said that although the firm ultimately shut down the operation, at least a handful of the attacks succeeded in stealing sensitive information."
A team at Anthropic uncovered an elaborate cyber-espionage scheme in which hackers, strongly suspected to work for the Chinese government, targeted government agencies and large corporations worldwide. The attackers used Claude Code's agentic capabilities and attached external tools like password crackers to analyze vulnerabilities, write malicious code, harvest passwords, and exfiltrate data. Claude performed extended autonomous actions for hours while human operators briefly reviewed results and triggered next steps. The operation followed a professional schedule aligned with the Chinese workday and holidays. Anthropic shut down the operation, but several attacks succeeded in stealing sensitive information.
Read at The Atlantic
Unable to calculate read time
Collection
[
|
...
]