Security researchers have uncovered what they believe to be the first documented case of an AI agent autonomously executing a ransomware attack from start to finish.
AI Just Pulled Off Its First Solo Cyber Attack — and Experts Are Alarmed
A cloud security team at Sysdig discovered the AI attacker, which they named Jadepuffer, broke into a vulnerable server, discovered login credentials, encrypted a production database and demanded a bitcoin ransom — all without human assistance, The Independent reports.
Michael Clark, Sysdig’s director of threat research, described it as “the first documented case of agentic ransomware: a complete extortion operation driven end-to-end by a large language model.” The AI adapted its tactics in real time, going from a failed login attempt to a working fix in just 31 seconds.
Researchers also noted that even if victims paid the ransom, they could not recover their data — the AI had already deleted it without creating a backup. The findings are yet to be independently verified, but follow a rare joint warning last month from the Five Eyes security alliance that AI is “months away” from wreaking havoc on businesses and governments.
NEWS NOW:
