The ‘first’ AI-run ransomware attack still needed a human
TechCrunch
Researchers found that while an AI agent executed a ransomware attack, a human still performed the initial setup and target selection.Summary
Security firm Sysdig recently documented a ransomware operation called JadePuffer, where an AI agent autonomously executed technical tasks such as exploiting vulnerabilities and encrypting files. Despite claims of no human oversight, Sysdig clarified that a human operator was still essential for provisioning infrastructure, obtaining initial credentials, and selecting victims. While the AI’s ability to adapt and narrate its own progress is notable, the attack highlights that autonomous agents remain tools requiring human direction rather than fully independent criminal entities.
(Source：TechCrunch)