Taiwan says AI agents used in cyberattacks targeting islandTaiwan says AI agents used in cyberattacks targeting island2 min readAugust 13, 2026 - 5:51PMAFPOverseas hackers used AI agents to help carry out a wave of cyberattacks targeting Taiwanese government agencies in July, the island's digital affairs ministry saidOverseas hackers used artificial intelligence agents to help carry out a wave of cyberattacks targeting Taiwanese government agencies in July, the island's digital affairs ministry said Thursday.Taiwan has previously blamed Chinese hackers for carrying out most of the millions of cyberattacks affecting the island democracy on a daily basis, but did not identify a particular country in this latest incident.China claims Taiwan is part of its territory and has intensified military, political and diplomatic pressure on the island in recent years. Taipei also accuses Beijing of using a range of "greyzone" methods -- tactics that fall short of an act of war -- including cyberattacks to harass the island.  The cyberattacks began in July and the "affected agencies have successively completed the response measures", the Ministry of Digital Affairs said in a statement, without providing further details on the scale, damage or targets of the attacks. Hackers used a "hybrid model" that combined AI agents, such as the open-source AI agent platform OpenClaw, to assist in the attacks, the ministry said.Others you may likenewsnews"AI agents can rapidly chain together multiple attack techniques and use secondary systems -- such as backup and testing systems -- as jumping-off points, giving the attacks the characteristics of high speed, low cost, and large scale."Chinese tech giants seeking to take advantage of the interest in OpenClaw have been offering simplified installation and affordable coding plans to help users host its agents on cloud servers.The digital affairs ministry's statement was in response to a Financial Times article this week that identified Taiwan as the target of an autonomous AI attack by suspected Chinese hackers. The FT, citing researchers at Israeli AI company Dream, said hackers had used open-source AI agents to build an autonomous tool to carry out the attack.The tool deployed up to eight autonomous agents that mapped 21 government systems, explored vulnerabilities and changed tactics when blocked, the FT said, describing the attack as a first of its kind.While AI cyberattacks are "not a new thing", this was "the first time it has been revealed that multiple AI agents were used to carry out a cyberattack," said Kenny Huang, chairman of Taiwan Network Information Center, an internet services company. Several countries, notably the United States, have voiced alarm in recent years at what they say is hacking activity backed by Beijing targeting their governments, militaries and businesses.China has denied the allegations, previously insisting it opposes and cracks down on all forms of cyberattacks.tpj-joy/amj/scoMore related storiesAFLSuns coach yet to unlock star recruitGold Coast coach Damien Hardwick has admitted he is yet to find the right way to use former Melbourne star Christian Petracca.Read moreBreaking NewsRescuers search for survivors as Indonesian quake death toll rises to 40Rescuers search for survivors as Indonesian quake death toll rises to 40Read moreNTFinal farewell for Kumanjayi Little BabyAlice Springs united in sorrow on Saturday as mourners dressed in pink farewelled Kumanjayi Little Baby.Read more