Virginia-based data labeling and AI startup Enabled Intelligence is expanding its repository of curated datasets used by government and commercial partners to train and deploy models, adding a new collection of drone footage recorded in war-torn Ukraine.
Since Russia’s large-scale invasion of its territory in 2022, Ukraine has produced an astonishing amount of frontline video depicting real-world combat operations.
Steadily increasing visual recordings have created vast amounts of training data, allowing military observers and defense contractors to rapidly innovate weapon capabilities based on the latest tactical lessons, such as AI models that allow drones to autonomously recognize and attack targets.
“This is the first Ukrainian full-motion video in our EView library,” company CEO and founder Peter Kant told DefenseScoop on Monday. “What’s different about this is that it’s real, not a simulation or a controlled environment.”
Its data library includes labeled datasets across multiple sensor types, including electro-optical, synthetic aperture radar, infrared, and foreign language audio.
Kant said “over 500,000 hours of Ukraine conflict drone footage” will make up the new collection. The images are pre-labeled, validated, and considered ready for use in training AI models.
“This is the most complex and dynamic conflict footage in modern history, labeled across aerial object detection, vehicle classification, and ground operations. That kind of operational reliability is very difficult to replicate and is exactly what AI systems need to perform when deployed,” Kant said.
Founded in 2020, Enabled Intelligence specializes in data acquisition and conditioning, precision labeling and annotation, and engineering and deployment of high-performance custom models for certified end users in the U.S. military and government, as well as the healthcare, financial services, and energy industries.
The new dataset has a variety of use cases, but the CEO expects it to be particularly valuable for those looking to reduce the need for labeling and validation, and speed up the training time of AI models for aviation applications. Kant says the data will allow different types of drones to be “AI-enabled right away.”
“Ukraine has produced more real-world drone footage than any other conflict in history,” he said. “That data is only valuable if someone takes the hard work of making it usable.”
On the commercial side, Kant cited product delivery and remote sensing as potential drone-related applications in the near future.
“In defense, this data is used for intelligence gathering, offensive and defensive operations, and even logistics, including moving supplies into conflict zones that are too dangerous to send people to,” he told DefenseScoop.
The National Geospatial-Intelligence Agency has hired Enabled Intelligence to award a single award contract worth up to $708 million over a seven-year award period in 2025. The Sequoia data-labeling-as-a-service contract broadly includes work to train computer vision algorithms used in intelligence, surveillance, and reconnaissance, or ISR, operations.
This data labeling effort is the basis, among other things, for NGA’s Maven program component.
Kant did not provide details about the government agencies and defense customers the company currently works with, or the source of the Ukrainian full-motion video drone assets in the company’s data library.
But he said the new collection is “ready and available now to authorized users working in the United States, Ukraine, and NATO allies.”
“What makes the footage from Ukraine particularly valuable is that it’s real. You get all the weather conditions, all the terrain types, all the unpredictable scenarios that simulations can’t reproduce,” Kant told DefenseScoop.
