AI-Run Ransomware Attack Still Depended on Human Setup, Researchers Say

The first documented "agentic" ransomware attack used an autonomous AI agent for technical execution, but a human provisioned infrastructure and chose the victim.

• Sysdig researchers documented the first known case of "agentic ransomware," dubbed JadePuffer, in which an AI agent broke into a server, encrypted files, and wrote its own ransom note.
• Sysdig senior director Michael Clark said a human still set up the operation, provisioned the command-and-control server, and provided credentials obtained through a prior compromise.
• The AI agent exploited known vulnerabilities in Langflow and MySQL to gain admin access and encrypted over 1,300 configuration records.
• Clark said Sysdig could not identify the specific AI model driving the agent, but he expects more such attacks given the low cost of running agents.