Eight agents, four days, 21 government systems: the first hands-off attack on a state
The Israeli security firm Dream published research on 12 August describing a four-day intrusion in early July against Taiwanese government targets that it says ran largely without human steering. The attackers assembled their platform from two publicly available agent frameworks, Hermes and OpenClaw, and ran as many as eight agents at once. The system mapped 21 government systems, searched vulnerability databases, GitHub and published security research for techniques worth trying, scanned targets in parallel and changed tactics when it was blocked. It compromised at least 85 government accounts and extracted more than 2,500 personnel records before moving on to IT suppliers, the country's nuclear safety agency and at least seven energy companies. Dream found the operation in an archived 160-megabyte collection of nearly 1,400 files and attributes it to China-linked operators. Humans were still needed to set up the agent coordination and the decision logic.
Related stories
DeepSeek's cheap workhorse can now see — and on agent tasks that need eyes it says it is close to Anthropic's best2026-08-21Anthropic expects to raise as much as SpaceX did — the largest listing in history — and could file publicly this month
2026-08-21Nvidia is paying $6 billion for the machine that builds a rival's models — and hiring 109 of the people who ran it
2026-08-21The doctors and lawyers who grade AI answers are now a half-billion-dollar business — and this one grew fivefold in eight months
2026-08-21Given four hours and a GPU to improve the way AI is trained, the best agent scored 0.25 out of 1 — and most never tried
2026-08-21