For now, most of the drone wars in Ukraine are manipulated. We see a glimpse into the future, but we are not there yet.
Ukraine and Russia are experimenting with artificial intelligence-enabled drones, but full autonomy and full-scale deployment remain limited in combat, researchers say.
AI-enabled autonomy in uncondemned systems can reduce the burden on human operators of how drones are used on the battlefield, bypassing electronic warfare and signal jamming, and speeding up targeting and decision-making processes. It also allows you to analyze and adapt data in real time. This has an advantage in combat.
Not autonomous and not used at large scale
Katerina Bondar, a fellow at Wadwani AI Centre at the Center for Strategic and International Studies, reported in March that autonomy, the ability of a system to operate independently in complex environments with limited supervision, “does not yet exist on the battlefield of war in Ukraine.”
A new report from the war expert Katerina Stepanenko Institute explains, “Neither Russia nor Ukraine had used AI/ML drones on the battlefield as of early June 2025.”
Both Russia and Ukraine face the challenge of continuing to innovate AI and machine learning with drones. AP Photo/evgeniy maloletka
The report states that both parties “increasingly integrate ML capabilities and consolidate limited AI adaptation to new drone variants on the path to developing fully AI/ML-powered drones.”
Drones that use machine learning require most of the involvement of operators, even some AI. Machine learning allows drones to perform pre-programmed tasks, but lack the autonomy provided by true AI capabilities.
At least it's not an AI revolution yet
The Ukrainian military observed a Russian flight drone swarm that relied on visual terrain navigation using onboard cameras and maps instead of vulnerable GPS, and observed that it was able to autonomously detect and select targets without operator input.
Later last month, Ukraine said it had deployed an AI-enabled “Mother Drones” that could autonomously send first-person view drones to hit targets. And Ukrainian security services reported earlier this month that the rude system used in the operation of spider web attacks on Russian aircraft on military aerial floors had switched to using AI to complete the mission if it lost a signal with the operator. Security Services said this was done using AI algorithms and manual operator interventions.
A Bondar report on Drones in March said that AI-equipped attack systems are three to four times more likely to reach targets than drones piloted by humans alone.
But has the revolution come?
One commander with another drone unit said last fall he hopes for an AI-enabled drone that doesn't require pilots to be on the battlefield within six months. That future is not here yet. Earlier this year, frontline Ukrainian drone units told Business Insider that AI-enabled drones are not yet widely used.
War is the proven foundation of cheap drones and emerging technologies. However, turning a prototype into a scalable, battlefield capable AI fleet requires data, chips and adjustments that neither side currently has.
AI-enabled drones ease some of their dependence on human operators and can complete many tasks on their own. Wolfgang Schwan/Anadolu Agency via Getty Images
For Russia, Stepanenko relies on further development of these critical combat capabilities to rely on the collection, storage and management of battlefield data to train mission AI and critically organize the best ways to identify enemy drones from the friendly ones.
Ukraine is already working on the latter with situational awareness systems such as the Delta and Klopiva. This reports Stefanenko is similar to the command and control system the US Department of Defense envisions.
For example, Delta provided Ukrainian troops throughout the branch and coordinated intelligence from a variety of systems, including drones, sensors, frontline reconnaissance and satellites.
However, in trying to overcome the challenges of wider development, Russia is fighting for drone innovation and centralisation of production under its government, which could hinder progress. Meanwhile, Ukraine is struggling with resources. Ukraine also faces problems with government coordination, computing power and lack of maintenance.
Ukrainian drone operators say the full-scale deployment of AI- and ML-driven drones will completely change the drone war. 24th Mechanized Brigade via Oleg Petrasiuk/Ukraine's AP
Among these challenges, Ukrainian drone developers have nevertheless become models for other worlds. Companies are working closely with frontline forces to meet their needs and effectively create relatively low-cost systems that push envelopes up to new features such as drone swarm technology.
But in the meantime, more testing and investment will be required by both sides of this war before autonomous, AI-enabled systems actually mark the battlefield.
However, the expected changes they could bring will overhaul how drone wars will be fought. Ukrainian typhoon drone units told BI that if they become prolific on the battlefield, they will completely change the way operators use drones for reconnaissance and strike missions to oppose enemy personnel, positions, equipment and aviation targets.
