Taiwan\u2019s statement comes a day after reports that suspected China-linked hackers had carried out a first-of-a-kind breach<\/p>\n
Taiwan says it detected AI-assisted cyber-attacks on government agencies that came from overseas last month, a new kind of threat that has been reported as \u201cfirst-of-a-kind breach\u201d.<\/p>\n
The Ministry of Digital Affairs (MDA) said its cybersecurity monitoring units \u200cdetected the \u201cabnormal attack\u201d targeting government agencies, which began on 20 July. The National Institute of Cyber Security issued a series of warning alerts while it investigated.<\/p>\n
The Financial Times quoted Dream \u2013 an Israeli AI company that detected the intrusion \u2013 as saying the attackers used open-source AI agents to build an autonomous hacking tool that behaved like a coordinated cyber team, describing it as a first-of-a-kind breach.<\/p>\n
Dream was reported as saying that the tool compromised \u201cat least 85 government user accounts, extracting more than 2,500 personnel records before expanding the attack to Taiwan\u2019s nuclear safety agency and at least seven energy companies\u201d.<\/p>\n
Taiwan has in recent years complained about what it sees as China\u2019s \u201chybrid warfare\u201d \u2013 from daily military drills near the island to disinformation campaigns and cyber-attacks \u2013 as Beijing ramps up military and political pressure \u2060on the democratically governed island to force Taipei to accept its claims of sovereignty.<\/p>\n
Though Taiwanese officials did not direct accusations at China, the FT on Wednesday reported that China-linked hackers were suspected. Dream did not attribute the attack to a specific group, but said the use of Simplified Chinese in internal communications linked to the hack meant there was a high probability the operator was connected to China, the FT reported.<\/p>\n
Chinese cyber-attacks on Taiwan\u2019s key infrastructure, from hospitals to banks, rose 6% in 2025 from the previous year to an average of 2.63m attacks a \u2060day, the island\u2019s National Security Bureau said in January, adding some hacks were synchronised with military drills in \u201chybrid threats\u201d to paralyse the island.<\/p>\n
The investigation into the AI-assisted attack said it showed clear characteristics of an \u201coverseas source\u201d, with hackers employing a hybrid approach that combined manual operations with AI agent-assisted attacks, such as Open Claw, the MDA said.<\/p>\n
\u201cThe relevant attack sources, methods, and scope of impact have all been fully investigated, and the affected units have successively completed their handling,\u201d the ministry added.<\/p>\n
In response to this new type of threat, the government has established protective guidelines and strengthened system monitoring to block attacks early, it said.<\/p>\n
The statement did not mention China, and China\u2019s Taiwan Affairs Office did not immediately respond to a request \u200cfor comment on the attack.<\/p>\n
Taiwan\u2019s MDA and Chinese authorities did not comment on the earlier FT report.<\/p>\n
The threat of AI-assisted hacking campaigns has \u200bbeen growing for years but ramped up dramatically \u200bover the past few months after top AI labs \u200breleased models, such as Anthropic\u2019s Mythos, which can rapidly conduct reconnaissance, identify vulnerabilities, and take advantage of them.<\/p>\n
Cris Thomas, a \u200bresearcher who works as a \u200csecurity advocate at the code \u200bsecurity company Semgrep, said the \u200bspy campaign was an illustration of just how quickly AI-assisted hackers could hit their mark.<\/p>\n