AI-driven cyberattack on Taiwan government systems
Taiwan's government agencies were targeted by foreign hackers using autonomous AI agents to conduct cyberattacks in July 2026, reportedly stealing over 2,500 personnel records and compromising 85 government user accounts. Israeli cybersecurity researchers reported the incident involved AI agents operating independently to map systems, conduct reconnaissance, and coordinate attacks without human intervention.
What happened
Taiwan's Digital Development Department disclosed on 13 August that Taiwan government agencies suffered AI-assisted cyberattacks from foreign hackers. Security monitoring units detected abnormal attacks against government agencies in July, triggering investigation and response.
Israeli cybersecurity company Dream reported discovering AI-driven hacking activity in which autonomous AI agents mapped 21 government systems, breached 85 government user accounts, and stole over 2,500 personnel records within 4 days in July. The attacks also extended to Taiwan's nuclear safety authority and at least 7 energy companies.
Reported impact
More than 2,500 personnel records were stolen from Taiwan government systems. The attacks expanded to Taiwan's nuclear safety authority and at least 7 energy companies.
Taiwan's Digital Development Department stated that affected units had completed processing and handling of the incidents.
Index analysis
The attacks employed AI agents, including Open Claw, in a hybrid mode that used backup and test systems as stepping stones, enabling attacks with high speed, low cost, and large scale. The AI system operated autonomously, conducting reconnaissance, credential attacks, and planning subsequent attack paths without human intervention, coordinating up to 8 AI agents.
Taiwan's National Security Bureau reported in January 2025 that cyberattacks on Taiwan's critical infrastructure increased 6% year-on-year, averaging 2.63 million attacks daily. Internal documents associated with the hacking contained simplified Chinese characters, suggesting the perpetrators were likely government or civilian organizations connected to China, though Taiwan government and Dream did not confirm the source.
Governance controls
Taiwan's Digital Development Department stated it would strengthen overall government defense capabilities based on attack characteristics identified in the investigation and continue monitoring for other potential attack paths. The government issued defensive guidelines for such cybersecurity threats.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.