AI Agents Conduct First Fully Automated Cyberattack on Taiwan Government Systems
Summary
AI agents carried out a fully automated cyberattack on Taiwan government systems, cracking 85 user accounts and extracting records of 2,500 employees. Taiwan’s Digital Affairs Ministry said the intrusion came from abroad and used a hybrid approach combining traditional hacking with AI tools.
Key Points
- Hackers used automated AI systems to conduct what experts say is the first publicly disclosed fully automated cyberattack on government agencies.
- The attack mapped 21 government systems in four days, cracked 85 accounts, and extracted records of 2,500 employees.
- Taiwan’s Digital Affairs Ministry said the attack originated from abroad and used a hybrid approach combining traditional hacking with AI agents like OpenCL.
- AI company Dream said the hackers’ system used open-source AI agents to automate coordination, adapt to blocks, and lower the cost of hacking.
