Russia and Ukraine Accelerate Push Toward Autonomous Battlefield Warfare

Tetiana Bubynets was thrown onto the road after a Russian drone detonated at a gas station in the southern Ukrainian city of Zaporizhzhia.

Although the July 6 attack sent a violent shockwave through the area and showered it with drone fragments, the 18-year-old accounting student had no visible injuries beyond a nosebleed.

The strike and its aftermath

Bubynets was taken to hospital, where she was pronounced dead shortly after arrival. Two other civilians, aged 41 and 48, also died in the attack.

Tetiana Melikhova, one of Bubynets’s lecturers at Zaporizhzhia National University, said the student was an only child and that her mother “lost the meaning of life” after her death.

“She couldn’t speak. At the funeral, she just lay on the coffin; she couldn’t let her child go,” Melikhova told Al Jazeera.

According to a New York Times analysis, the strike that killed Bubynets was the first documented incident involving Russia’s use of a fully autonomous attack drone powered by artificial intelligence.

Ukraine has endured repeated drone attacks since Russia launched its full-scale invasion in February 2022. However, Ukrainian officials who inspected the wreckage said the unmanned aircraft differed from other drones used in the conflict.

Instead of an antenna, the drone contained a commercially available mini-computer made by US technology company Nvidia. The system could identify the type of target it had been programmed to detect—apparently propane tanks in this case—lock on and continue toward the target without further human control.

Analysts say autonomous drones are only one component of a wider race between Russia and Ukraine to expand the technology infrastructure used for reconnaissance, targeting and strikes.

“It’s not just about a drone having AI in it, seeing something and striking,” said Marina Miron, a researcher in the defence studies department at King’s College London. “Rather, it’s about the drone detecting something, communicating it to a software system that can map the GPS coordinates onto a live battlefield map and notify the nearest artillery unit.”

Known as reconnaissance-strike architecture, these integrated battlefield management systems function like a central command. They monitor the positions of friendly forces while identifying likely enemy locations.

AI accelerates the process by rapidly analysing enormous volumes of information gathered from drone sensors and other military systems, including interceptor missiles and artillery units.

The resulting tactical advantage comes from compressing the “kill chain”—the process of coordinating intelligence and transmitting it to forces on the ground so targets can be found, tracked and destroyed within minutes.

Within this network, autonomous AI-enabled drones are merely one part of a much larger system.

Ukraine is developing its own platform, known as DELTA. After operational trials in late 2025, Russia began deploying its equivalent—the Svod Tactical Situational Awareness Complex—to front-line units.

Miron said both sides maintain that, after battlefield data has been processed, a decision-support system presents commanders with a range of options, thereby keeping a person in the decision-making loop.

“How far that is true is, of course, impossible to find out,” she added.

Admitting that such systems possess full AI autonomy would raise significant ethical and legal concerns.

Even when humans retain formal authority to approve attacks, research suggests that AI can subtly influence their choices and effectively predetermine the outcome.

“Under time and pressure, commanders often do as AI suggests,” said Miron, whose work focuses on military AI and remotely piloted and autonomous systems.

The path to full automation

Coordinated AI-enabled military systems represent the next stage of modern warfare, according to Serhiy Beskrestnov, a Ukrainian Ministry of Defence adviser on loitering munitions known by his callsign Flash.

“That’s what it will all come to,” he wrote on his Telegram channel on July 1. “Navigation, search for a target, and attacks will be fully autonomous.”

Drones with limited autonomous functions are already being deployed on a large scale.

Heorhii Volkov, commander of Ukraine’s Yasni Ochi drone strike unit, said they are currently used mainly during the final phase of long-range missions. In those operations, first-person view drones may lose their radio link because of electronic warfare or the curvature of the Earth.

AI-enabled autonomy addresses this problem during what are known as “last-mile” operations. If communication is interrupted, an onboard AI module assumes control, locks onto the target and completes the strike without further human input.

Yaroslav Azhnyuk, founder of the Ukrainian technology company The Fourth Law, which develops autonomy software for unmanned vehicles, said Russia and Ukraine are working to field a broader range of autonomous systems.

“This is not simply a race to demonstrate a fully autonomous drone. It is a race to make autonomy reliable, affordable and deployable at scale,” Azhnyuk told Al Jazeera. “Electronic warfare, increasingly long engagement distances and the growing number of drones are making human-controlled systems less sufficient. Autonomy solves exactly these constraints.”

Azhnyuk believes autonomy could become a decisive factor within the next six to 12 months.

“The decisive advantage will belong to the side that can deploy reliable autonomy across tens or hundreds of thousands of systems, not just demonstrate it in prototypes,” he said.

Kateryna Bondar, a senior fellow at the Center for Strategic and International Studies, highlighted the risks created by AI’s dependence on algorithms.

“The system doesn’t understand what it’s doing; it’s just following the algorithm and the rules which people created,” said Bondar, who works at the CSIS’s Wadhwani AI Center. “AI has to be trained to recognise targets and engage these targets, but writing code that distinguishes between a bus and a tank is easier than writing code that distinguishes between a Russian and a Ukrainian soldier.”

Many inexpensive first-person view drones also use low-quality cameras, and fog, smoke or other adverse atmospheric conditions can further reduce image resolution.

Bondar said military units operating autonomous systems therefore “delegate the decision to a very imperfect technology that cannot distinguish between a civilian and a military person, combatant or non-combatant”.

“This is why there is a conversation around who makes a decision and who has responsibility for mistakes and errors.”

Source link