First AI Agent Ransomware Strikes Autonomously in Seconds

Researchers have documented the first known ransomware operation powered entirely by an AI agent, capable of executing attacks from start to finish without human intervention during the intrusion. The system, dubbed JadePuffer by Sysdig's Threat Research Team (TRT), uses a large language model (LLM) to reason, adapt, and complete tasks in seconds. Source: elespanolSource: infobae

How it works

While a human still selects the target and sets up infrastructure, the AI agent takes over once the attack begins. It uses natural language reasoning to prioritize objectives, document steps, and retry failed attempts with refined parameters. In one case, it went from a failed login to a working solution in 31 seconds. Source: elespanol

Implications

Experts warn that such automation could lead to a surge in cyberattacks, lowering the skill and cost barriers for criminals. The discovery follows concern over powerful models like Anthropic's Mythos, which was briefly banned by the US government. However, Doris Seedorf of Softtek notes this is an intensification of existing risks, not a radical change. Source: infobae