TL;DR
Ukraine’s Defense Ministry is using Avengers Labs to give controlled access to annotated combat-drone data for AI training. Companies work inside a protected Dataroom and return improved models to Kyiv, while the size, performance gains and foreign participant list remain partly opaque.
Ukraine’s Ministry of Defense is using Avengers Labs to let Ukrainian and allied defense companies train computer-vision models on millions of annotated frames from real combat drone missions, while requiring that improved models be returned to Kyiv. The arrangement matters because battle-tested data from drones operating under electronic warfare, bad weather and hostile fire is one of the scarcest inputs in military AI.
The program runs inside Brave1, Ukraine’s defense-innovation cluster, and uses a protected Dataroom built by the Defense Ministry, the Ministry of Digital Transformation, the Armed Forces, a military-intelligence research institute and Palantir, according to Ukrainian officials cited in the source material. Companies do not receive raw front-line video; they work with structured visual and thermal imagery in the secure environment.
The datasets cover aerial and ground targets recorded under operational conditions, including night scenes, fog, rain, camouflaged vehicles, drones in flight and targets captured by different sensors. Ukrainian officials say millions of frames have been annotated from tens of thousands of drone sorties. Those figures are based on official statements and were not independently audited in the supplied material.
More than 100 Ukrainian companies are described as having Dataroom access, while international developers participate through Avengers Labs. The Defense Ministry’s Avengers platform, developed by its Innovation Center, is already used to detect, classify and track hostile targets from drone and camera feeds; the ministry says it flags about 12,000 enemy units per week and feeds the VEZHA streaming module inside the DELTA battlefield-management system.
Avengers Labs
Ukraine’s Ministry of Defense is renting access to the world’s only large-scale, real-war computer-vision dataset. The terms: train your model inside the protected Dataroom — Ukraine keeps the finished AI.
Inside the Dataroom
- Structured visual & thermal imagery of aerial and ground targets
- Hard cases: camouflaged armor, night, fog, rain, multiple sensors
- Feeds the Avengers platform inside the DELTA / VEZHA system
- Focus track: automatic detection & interception of enemy drones
The goal
- 100% of frontline drones with onboard machine vision
- Autonomous navigation in GPS-denied / jammed (EW) skies
- Autonomous Shahed interception — human keeps the trigger
- Scaling vs. Shahed launches rising ~35% / month
Combat Data Becomes Leverage
For defense AI developers, the value is not only model architecture. It is labeled data showing hard cases: jamming, darkness, camouflage, weather and fast-moving targets. Ukraine is offering access to that data without handing over the archive itself.
For Kyiv, the exchange turns wartime footage into strategic leverage. Companies get rare training material, while Ukraine receives models that can be tested against Russian tactics. The program also shows a shift in military procurement: operational data is becoming part of the bargain, not just a byproduct of combat.
AI training datasets for drone surveillance
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From Army Of Drones To Avengers
Ukraine has spent the war building a fast defense-tech pipeline around drones, software and battlefield feedback. Brave1, created in 2023 by Ukrainian government bodies, was designed to connect developers with military needs and move usable tools toward the front faster.
Avengers sits within that ecosystem as a computer-vision platform tied to DELTA, Ukraine's situational-awareness system. The new partnership layer adds a marketplace-like exchange around the training data behind those systems. The focus tracks cited in the source material include onboard machine vision for frontline drones, navigation in GPS-denied or jammed skies, and automated detection for Shahed interception, with a human retaining control over firing decisions.
"win-win"
— Ukraine's Defense Ministry framing
Autel Robotics EVO II Dual 640T V3, 640x512 Thermal Imaging Drone, Remote ID Supported, 0.8'' CMOS&50MP 8K RYYB Sensor,10+ Temp Measurement, D-RI Ranges, Thermal Analysis Tool, for Search&Rescue/Power Inspection/Public Safety/Hunting
🏆【AUTEL ROBOTICS USA AUTHORIZED】We are the longest and largest distributors of Autel Robotics. Over the years, we have...
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Model Performance Still Unproven
Several points remain unresolved. Ukraine has not made the raw dataset public, so outside experts cannot verify its size, label quality or representativeness. The exact list of foreign participants, commercial terms and security limits has also not been disclosed in the source material.
It is also not clear how quickly returned models will move from testing to front-line use, how failures will be measured, or how autonomy rules will be enforced. Ukrainian officials describe autonomous navigation and Shahed interception as focus areas, while saying a human keeps control over firing decisions.
Jetson Thor 128G Developer Kit AI Performance 2070 TFLOPS with SSD, AI Edge Computer for Autonomous Robots, LLM, Computer Vision
【AI Performance for Edge Computing】 Powered by N-VIDI-A Jetson AGX Thor module with 128GB memory and 2070 TFLOPS...
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Returned Models Face Field Tests
The next test is whether models trained inside the Dataroom improve detection and interception in conditions that Russian forces keep changing through jamming, decoys, weather and modified drones. Ukraine is expected to keep expanding access for domestic and allied developers while prioritizing onboard machine vision for frontline drones and counter-Shahed systems.
Readers should watch for named foreign partners, independently described performance gains and deployment data from units using returned models. Until then, Avengers Labs is best understood as an active defense-AI data program with major official claims and limited outside visibility.
TOPDON TC004 Mini Thermal Imaging Camera, 240 x 240 TISR Resolution, 15-Hour Battery Life, 128 x 128 IR Resolution, 25Hz Handheld Infrared Camera with High/Low Temp Alerts, Auto Shutdown- Blue
【Enhanced Thermal Clarity】Start with 128x128 thermal imaging and enhance to 240x240 resolution with TISR technology for greater details....
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is Avengers Labs?
Avengers Labs is a partnership platform run by Ukraine's Ministry of Defense inside the Brave1 defense-innovation cluster. It gives approved defense developers controlled access to annotated combat data for AI training.
What data are companies allowed to use?
Companies work with structured visual and thermal imagery from real drone missions, including aerial and ground targets recorded in difficult battlefield conditions.
Do companies receive raw combat footage?
No, according to the source material. The data is used inside a protected Dataroom, and the finished or improved AI models are returned to Ukraine.
Why is this data valuable for AI?
Battlefield computer vision needs examples of real targets under jamming, darkness, camouflage, weather and sensor variation. That kind of labeled data is hard to scrape or simulate.
Are the systems making firing decisions?
The source material says the focus includes detection, navigation and interception support, while a human keeps control over the trigger. How those rules are applied in future deployments remains a point to watch.
Source: Thorsten Meyer AI
