Suspected China-linked hackers used AI to run the first-ever end-to-end autonomous cyberattack on Taiwan’s government, Israeli firm says — open-source-built tool continuously devised effective hack strategies in real-time
Hackers with suspected links to China used publicly available AI tools to carry out what researchers describe as the first observed end-to-end autonomous cyberattack against a government target, compromising at least 85 user accounts and stealing more than 2,500 personnel records from Taiwanese government systems, according to an August 12 Financial Times report, citing researchers at Israeli cybersecurity company Dream. The researchers say the attackers assembled an autonomous hacking platform using open-source AI-agent frameworks, enabling multiple agents to simultaneously map networks, research vulnerabilities, attempt intrusions, and adapt tactics when an attack path failed.
The campaign reportedly ran for four days at the beginning of July and at times deployed as many as eight autonomous agents in parallel. Dream said the system mapped 21 government systems before compromising user accounts and extracting personnel information. The attackers subsequently expanded their activity to Taiwan’s nuclear safety agency, at least seven energy companies, government suppliers, and other government systems.
Dream says it found the evidence inside a 160-megabyte (160MB) online archive that surfaced during its broader tracking of cyberthreat actors. The archive reportedly held 1,395 files showing that the tool was built on two open-source AI agent systems — Hermes and OpenClaw — both of which can be downloaded freely and are designed to let large language models carry out multi-step tasks on their own.
Researchers could not determine which underlying model powered the agents, but the data reportedly showed the model’s safeguards had been sidestepped by presenting the intrusion as an authorized penetration test rather than a real attack. Of particular concern is that the toolkit for the hack comprised such easily available systems, neither of which was purpose-built for offense. The operators appear to have assembled a capable autonomous tool out of components any developer can pull down and run.
What the researchers describe as the tool’s most striking feature was its ability to continuously devise attacks on its own, rather than follow a preprogrammed route. The platform continuously assessed available evidence, ranked possible attack paths, and reprioritized them as circumstances changed. When one technique failed, the tool tasked another agent with searching the internet for information and developing an alternative approach.
Dream stopped short of attributing the campaign to a specific hacking group or country. However, the researchers said the operators’ internal communications were written in Simplified Chinese, suggesting what they called a high probability that the operator was connected to China. The company also declined to name the victim, citing policy, but confirmed that it had notified a country in the “Asia-Pacific” region. Also, the data pulled from the target was in Traditional Chinese — a script used on government sites in Taiwan, Hong Kong, and Macau. Financial Times said a person with knowledge of the incident identified the target as Taiwan.
The incident highlights a growing concern within both the cybersecurity and AI industries over what the latest AI models can do autonomously. Anthropic, OpenAI, and Meta have reported instances of new AI models launching unexpected cyberattacks during internal testing, an infamous example being the recent OpenAI agent’s attack on Hugging Face. In another instance, OpenClaw wiped the inbox of Meta’s AI Alignment director despite repeated commands to stop
Researchers have warned that AI agents are making it increasingly easy to automate portions of cyberattacks that previously required skilled human operators, another deadly feature in the era of AI hacking. Dream’s chief strategy officer, Amir Becker, warned that the arrival of such tooling used in the Taiwan attack means every government should now assume it is under permanent automated assault.
The risk is stark for Taiwan, which was already facing a staggering volume of cyberattacks before agents entered the picture. The island’s National Security Bureau reported in January that it faced an average of 2.6 million Chinese cyberattacks per day in 2025, up 6 percent from the previous year. Beijing claims Taiwan as part of its territory and has threatened to use force if necessary to bring the island under its control. According to the Financial Times, Taiwan’s Ministry of Digital Affairs declined to comment on the specific incident, citing confidentiality. However, a ministry spokesperson acknowledged that the integration of AI has transformed the nature of security incidents.
