'Near-autonomous' AI agents attack Taiwan's nuclear safety agency
Some say the world will end in fire, some say an agentic swarm
Chinese cyber operatives compromised Taiwan's nuclear safety agency, government systems, supply-chain vendors, and energy companies in what security researchers described as a near-autonomous attack. Over four days in July, AI agents infiltrated 85 government accounts, extracting over 2,500 personnel records. The attack framework, built on open-source AI tools, deployed up to eight sub-agents targeting 12 systems over 12 waves.
The operation began by mapping the government's entire ecosystem, extracting URLs, API endpoints, and authentication data. Agents discovered unsecured API endpoints, cracked passwords, and accessed internal information systems, exfiltrating sensitive data. The attack highlights the growing capability of AI agents to autonomously exploit vulnerabilities and potentially pose a threat to national security.
Written by urgent.news from The Register Science's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
