Russia has tested a new control and aiming unit for AI-powered turrets that can autonomously engage different types of aerial drones, according to Russia’s Tallamkho design bureau in Grozny.
The test results and early deployment plans come as drone warfare keeps evolving, with Ukrainian officials also describing new Russian drone capabilities and Ukraine launching a secure AI data platform aimed at faster counter-drone development.
What Russia says it tested
Tallamkho told TASS that a new turret control and aiming unit “successfully passed tests” and is designed to protect positions, vehicles, and ground robots.
The bureau said the unit can work in three modes: manual, semi-autonomous, and fully autonomous.
In the semi-autonomous mode described by Tallamkho, AI is used to acquire the target while an operator initiates firing.
In the fully autonomous mode described by the bureau, the turret operates independently.
Capability claims: drones and nighttime use
Tallamkho said the system is intended to engage “various types of aerial drones,” and it stated that the AI model it developed can recognize aircraft-type drones as well.
The bureau also said that, with certain configurations, the turret system can operate at night.
Tallamkho framed the autonomous option as a way to defend fixed positions “without human involvement,” arguing that such positions can be identified and targeted by artillery and that reducing human presence helps protect personnel.
Where the first prototype is going
Tallamkho said the first prototype of the new unit has been sent for testing in the “special military operation zone,” using Russia’s official term for the war in Ukraine.
The same report said the turret equipped with the new control unit is referred to as “Ronin-3,” and that it is adapted for installation on pickup trucks as well as on ground robotic systems.
Tallamkho also said a tracked or wheeled ground robot could conduct fire either under remote operator control or autonomously, including in scenarios it described as operating behind enemy lines.
The bureau added that the original concept focused on protecting positions from drones, while also saying it is developing AI models intended to operate against other types of targets.
Ukraine’s parallel push into AI counter-drone tools
Ukraine’s Ministry of Defence has unveiled what it described as a new secure data environment aimed at speeding the development of AI systems that can autonomously detect and intercept Russian unmanned aerial systems, according to National Security News.
That report said the initiative is called the “Brave1 Dataroom,” is part of the government-backed Brave1 defence innovation cluster, and uses software infrastructure provided by Palantir Technologies.
National Security News reported that the platform is meant to give Ukrainian developers access to structured visual and thermal datasets collected from frontline operations so machine-learning models can be trained for aerial threat identification, tracking, and neutralisation.
The same article said access is restricted to Ukrainian developers who pass mandatory security and compliance vetting, and that the ministry has opened an application process for qualified technologists.
Palantir executive vice president Louis Mosley was quoted as saying the initiative gives Ukrainian engineers access to advanced military software and data infrastructure, and he also said it could eventually help Ukraine share experience and capabilities with allies.
Ukrainian prosecutor describes “testing” AI-equipped drones over Kharkiv
In a separate report, TSN said Spartak Borysenko, identified as head of the Kharkiv Regional Prosecutor’s Office, stated that Ukrainian cities have “become a testing ground for Russian weapons,” including drones equipped with elements of artificial intelligence.
TSN reported that Borysenko described the drones as being in a testing phase and said they can be programmed to strike targets including crowds of people, specific vehicles, or buildings.
“If a drone detects an object matching its assigned target while en route, it independently decides to carry out the strike,” Borysenko said, as quoted by TSN.
TSN also reported that Borysenko cautioned that accuracy cannot be guaranteed, saying AI performance depends on factors such as image quality, frame rate, and resolution, among other technical factors.
