Russia and Ukraine Race Toward Fully Autonomous AI Warfare

An 18-year-old Ukrainian student, Tetiana Bubynets, was killed on July 6 when a Russian drone struck a gas station in Zaporizhzhia. While she suffered only a nosebleed from the blast, she later died in the hospital alongside two other civilians. Her lecturer at Zaporizhzhia National University, Tetiana Melikhova, described the profound grief of Bubynets’s mother, who “lost the meaning of life” after her only child died.

The attack is believed to be the first documented instance of Russia employing a fully autonomous AI-powered drone, according to a New York Times analysis. Ukrainian officials examining the wreckage noted the device lacked a traditional antenna, instead featuring a commercial Nvidia minicomputer. This hardware allowed the drone to independently navigate, identify its target—likely propane tanks—and execute the strike.

Experts describe these AI-enabled drones as part of a broader shift toward “reconnaissance-strike architecture.” Marina Miron, a researcher at King’s College London, explained that the technology goes beyond individual striking drones. Instead, it creates a unified battlefield management system where drones detect targets, share GPS coordinates via software complexes, and direct nearby artillery units to engage.

This architecture compresses the “kill chain,” reducing the time required to find, track, and destroy targets from hours to minutes. AI acts as an accelerator, processing vast sensor data from drones, missiles, and artillery in real time. Both Ukraine and Russia are advancing similar systems; Ukraine has trialed its DELTA platform, while Russia is deploying the Svod Tactical Situational Awareness Complex to front-line units.

While both sides claim human commanders retain final decision-making authority through decision support systems, Miron cautioned that AI subtly influences outcomes. Studies suggest that under pressure, commanders often follow AI recommendations, effectively ceding control to the algorithm.

Serhiy Beskrestnov, a Ukrainian Ministry of Defence adviser known as “Flash,” stated that navigation, target search, and attacks will eventually become fully autonomous. Already, Ukrainian forces utilize AI for “last-mile” operations, where First-Person View (FPV) drones lose radio connection due to electronic warfare or distance. In these scenarios, onboard AI takes over to complete the strike.

Yaroslav Azhnyuk, founder of the technology firm The Fourth Law, emphasized that the competition is not merely about demonstrating prototypes but about scaling reliable and affordable autonomy. He predicted that the decisive advantage in the coming six to twelve months will belong to the side that can deploy tens or hundreds of thousands of autonomous systems.

However, Kateryna Bondar of the Center for Strategic and International Studies warned of the dangers inherent in algorithmic decision-making. She noted that AI systems follow programmed rules without understanding context, making it difficult to distinguish between civilian and military targets, especially when using low-cost FPV drones with poor camera resolution. “Military units using autonomous systems therefore delegate the decision to a very imperfect technology,” Bondar said, highlighting the growing debate over accountability for errors in automated warfare.