Russian developers used Claude to build autonomous drone software, Anthropic reports
A small team of freelance developers in Russia used Anthropic's Claude AI to build software for an autonomous swarm of first-person-view "kamikaze" attack drones, according to a 154-page report from the company. The software included terminal guidance, target selection, and coordination between multiple aircraft, with the system programmed to pick targets and crash into them without human approval.
The finding underscores a growing concern for governments and security professionals: commercially available AI tools are being repurposed for military and cyber operations. Anthropic's report, which examined how bad actors used its AI system for state-sponsored surveillance and propaganda operations, found the freelance team repeatedly targeted a location in Ukraine's Donetsk region and used VPNs to bypass geographic restrictions on Claude.
AI used in cyberattacks against Ukrainian officials
Anthropic also found that cybercriminals and state-backed hackers are using AI to run large portions of cyberattacks. One hacking group allegedly ran phishing, hotel wifi hijacking, and WhatsApp-takeover operations against targets in the Ukrainian government, military, and diplomatic sectors, using AI at nearly every stage. The group's methods matched Russia-based threat actor Midnight Blizzard, which the US government has linked to Russia's SVR foreign intelligence service, according to Reuters.
The group built a system that automatically detected when its malware was flagged by security defenses and rewrote the code until it evaded detection. This kind of automated adaptation reduces the manual workload for attackers and speeds up the cycle between detection and counter-response. For professionals in AI for IT & Development, the report offers a concrete case study in how AI changes threat modeling and defensive planning.
Escalation on the ground in Ukraine
The report landed as Russian attacks on Kyiv entered a new phase of intensity. Jet-powered drones struck two petrol stations on Friday morning, killing at least two people and injuring 12, according to mayor Vitali Klitschko. A third strike hit another petrol station on Thursday. All three targets were operated by Ukrnafta, a state-owned oil and natural gas extraction company. A list of 18 Kyiv petrol stations expected to be hit by Russian drones had been circulating on Telegram since Wednesday.
Russia has for two weeks been carrying out near-constant attacks with faster jet-powered drones. Ukraine has also launched drone strikes on Russian territory, with local officials reporting two deaths in the Tula region overnight and one in the Kursk region on Friday. Three people were killed in Russian strikes in Dnipropetrovsk, Zaporizhzhia, and Kharkiv regions, according to regional governors.
Diplomatic and political context
Russian president Vladimir Putin blamed the West for the conflict, citing what he called constant pressure on Moscow and attempts to "drag" Ukraine into NATO. Speaking before the BRICS summit in India, Putin said "everything happening in Europe right now is basically the result of system errors by the western globalists in politics, security, and economics." He linked European voters' shift toward far-right parties to their understanding of the situation.
A Russian official will attend a G20 energy meeting in Houston next week, a White House official said. Last month, the US welcomed Russian finance minister Anton Siluanov to a G20 meeting in North Carolina, the first time Russia's finance minister attended the forum in person since the 2022 invasion. European Union efforts to reduce dependence on Russian oil and gas are faltering, with the bloc heading into winter with low gas stocks, according to EU auditors.
The US House of Representatives is expected to consider a Russian sanctions package next week. The bill passed the Senate with overwhelming support, but several top Democratic lawmakers worry it expands president Donald Trump's tariff authority in ways he could use against US allies.
Why this matters for government and IT professionals
For those working in government, IT, or communications, the Anthropic report is a signal that AI misuse is no longer speculative. Freelance developers with access to a commercial chatbot built autonomous targeting software. State-linked hackers automated malware evasion. The tools are general-purpose; the safeguards are being tested in real time. Anyone responsible for security policy, infrastructure, or public messaging should treat AI-assisted attacks as an operational reality, not an emerging risk.
Understanding how models like Claude are being exploited-and what controls exist-is now part of due diligence for teams evaluating AI adoption. Professionals can build that understanding through structured training such as Claude AI Courses & Certifications, which covers the technology behind the systems making headlines.
Your membership also unlocks:
