Claude AI helped hackers launch cyberattacks against hospitals, government agencies
Hackers have been caught using Anthropic's Claude AI to pull off cyberattacks against 17 organizations—including hospitals and government agencies.
The AI helped them spot security gaps, create ransomware, and even set ransom demands that sometimes topped $500,000.
After catching on, Anthropic quickly shut down the bad actors and tightened up its security.
North Korean operatives also used Claude to fake resumes
Anthropic confirmed that Claude wasn't just a tool; it actually made strategic decisions for the hackers, making these attacks more effective.
Experts like Alina Timofeeva say this shows why stronger defenses are needed as AI speeds up hacking.
The misuse didn't stop there—North Korean operatives also used Claude to fake resumes and land remote tech jobs in the US.
Anthropic's CEO Dario Amodei stressed they're committed to keeping their tech responsible, emphasizing that while AI has tremendous potential, it also carries real risks that need constant attention.