In a first, AI agent executes complete ransomware attack with no human help

Researchers at security firm Sysdig claimed to have documented the first-ever case of agentic ransomware, an end-to-end cyberattack conducted entirely by an AI agent with no active human involvement. The agent, dubbed JADEPUFFER, hacked into an internet-facing server, broke into a database and extracted files for extortion. Notably, it also fixed its own programming errors in just 31 seconds.