# AI Killer Drones Are 'Making Decisions to Strike': Russia Just Used One And 3 Ukrainian Civilians Died > Source: > Published: 2026-08-25 14:44:25+00:00 # AI Killer Drones Are 'Making Decisions to Strike': Russia Just Used One And 3 Ukrainian Civilians Died ## An attack drone, an AI computer and no human pilot at the final stage defines a new era of war A Russian [attack drone](https://www.ibtimes.co.uk/belgorod-drone-attack-five-dead-dozens-injured-1813198) using AI to select its target has killed three civilians in Ukraine, in what experts describe could be the first documented case of a Russian self-targeting drone causing civilian deaths. The 6 July strike in Zaporizhzhia has intensified fears over machines autonomously making life-or-death targeting decisions. The victims were 19-year-old university student Tetiana Bubynets, 41-year-old Oleksiy Svirin, and 48-year-old Roman Karpiy. Russian operators reportedly programmed the drone to fly toward a gas station, where its onboard system selected a target, most likely propane tanks. Before reaching them, however, the drone crashed into a wall and exploded, killing all three civilians. ## The Drone Was Not Simply Remote-Controlled According to Ukrainian officials and forensic investigators, the weapon was a Russian Molniya drone fitted with an NVIDIA Jetson Orin minicomputer. Unlike systems where a human selects the target before AI takes over during the final flight, the Zaporizhzhia drone reportedly selected its target itself. Ukrainian investigators said its cameras and onboard software could identify objects and choose what to strike once it reached the target area. The recovered drones reportedly had no radio antennas, while investigators examined an unencrypted computer containing terrain imagery and code showing which objects the system had been trained to recognise and attack. ## Machines Are Making Decisions to Strike For Ukrainian air-defence commander Col. Serhiy Minaiev, the development signals a frightening shift in warfare. 'In a few years, we will be living in a "Terminator" movie,' Minaiev said. 'It's no joke. Machines are making decisions to strike.' The bigger concern is that software could increasingly take over the human judgement traditionally made just before an attack. That could make weapons faster and harder to disrupt, raising questions over mistakes, accountability, and international humanitarian law. The technology at the heart of the investigation is surprisingly ordinary. NVIDIA confirmed that photos of the recovered component showed a Jetson Orin minicomputer, which is designed for AI, robotics, and computer-vision applications. The company said that the modules are consumer-grade products 'sold to students, developers, and start-ups for a wide range of beneficial applications.' While the company does not sell them in Russia, the hardware can reportedly be obtained through resellers. ## The Global Push Toward Autonomous Warfare Russia is not alone in developing self-targeting AI. Former Ukrainian Defence Minister Mykhailo Fedorov said Ukraine had tested a fully autonomous AI system in Russian-occupied Crimea, with no civilians reportedly killed in the tests. Concerns over [AI-enabled warfare](https://www.ibtimes.co.uk/chinese-ai-drone-warfare-1800287) extend beyond Ukraine. Israel's war in Gaza has been described as an 'AI human laboratory' amid its reported use of AI-enabled surveillance and targeting systems. In September 2025, the UN Independent International Commission of Inquiry concluded that Israel had committed genocide against Palestinians in Gaza, a finding Israel rejected. In the US, the [Trump](https://www.ibtimes.co.uk/trump-aide-natalie-harp-security-clearance-controversy-1815562) administration also clashed publicly with Anthropic over military AI. Anthropic refused to allow its systems to be used in fully autonomous weapons that remove humans from target selection and engagement. ## Rights Groups Warn of AI's Growing Role in War Human rights groups warn that AI could accelerate warfare and create risks beyond technical failures. Amnesty International and Access Now said AI could 'fundamentally transform military operations,' thus calling for an end to its use in military kill chains. The groups said they were 'profoundly concerned' that AI language models used to generate and prioritise targets could make it harder for militaries to uphold international humanitarian law. They pointed to the technology's 'speed and scale,' along with unreliable and biased input data, as major challenges to respecting the principles of distinction, proportionality, and precaution. © Copyright IBTimes 2026. All rights reserved.