Anthropic says Claude used for cyberattacks and surveillance
Anthropic said Russian- and Chinese-speaking operators used Claude to automate cyberattacks and vulnerability research. A Russian-speaking actor, “JackPoterz,” used customized AI workflows to streamline much of the attack chain against more than 20 targets, including government ministries, intelligence bodies, embassies, and diplomatic missions in Ukraine and Europe. Chinese-speaking operators used Claude to help with vulnerability research and orchestration, with one workflow reportedly generating more than a dozen possible zero-day findings in network-appliance firmware in a month. Anthropic said AI is lowering the cost and time of attacks, letting solo operators complete breaches in 2–3 hours and run multiple victims at once. Separately, a likely Bamako-based consultant working with Mali’s state intelligence service used Claude to help build a population-scale surveillance system covering about 25 million SIM cards, designed to produce intelligence dossiers without court orders.
