Ukraine and NATO have jointly developed two AI-based decision-support tools for the battlefield. One forecasts enemy drone activity and the other plans artillery fire. According to hartpunkt, which reported the development on October 16, 2026, Ukraine’s Ministry of Defence announced the systems on October 13 under the designations “Model No. 13” and “Model No. 32.” NATO’s Allied Command Transformation (ACT) worked with the Ukrainian armed forces on the effort, supported by the NATO-Ukraine Joint Analysis, Training and Education Centre (JATEC).
Model No. 13 is designed to predict enemy drone use along the front line. It combines historical operational data, weather conditions and the current tactical situation to forecast in which sectors and at what times the enemy will use drones, how intensively, and which types it is likely to field. The ministry says the model has been tested and refined against real datasets and is ready for integration into a Ukrainian battlefield management system, which hartpunkt judges is likely to be DELTA. The ministry gave no figures on forecast accuracy, data volumes or operational results (hartpunkt).
The goal is to give commanders earlier warning so they can adjust protective measures, reposition units and prepare active defenses. Hartpunkt stresses that the tool supports decisions but does not replace current reconnaissance and assessment of the situation (hartpunkt).
Model No. 32 automates much of artillery fire planning. It takes into account available weapons, ammunition stocks, firing positions and operational constraints, first filters out options that are not feasible logistically, ballistically or operationally, and then ranks the remaining options. The aim is to reduce the manual work of comparing fire plans and to shorten the time between identifying a target and executing a fire decision. A prototype has been handed to Ukrainian units for evaluation, and JATEC says wider rollout depends on further testing and validation (hartpunkt).
JATEC, which began operating in 2025, exists to analyze Ukrainian combat experience and share it with the Alliance. It confirmed the details on October 14, noting that both models were presented in September at a seminar on IT, AI and machine learning held at Lithuania’s National Cyber Security Centre in Vilnius, with around 30 experts from NATO bodies and the Ukrainian armed forces (hartpunkt).
The significance lies in how lessons now travel between Ukraine and NATO. Ukraine is no longer simply a recipient of allied equipment and training. Its combat data is being turned, with NATO, into tools the Alliance can learn from. Both models target the core of the current fight: drones and fires, and the speed at which forces can sense, decide and act.
The caution is in what has not been disclosed. Without published accuracy figures or combat results, these remain promising tools rather than proven ones. Their real test is whether they hold up against an adversary that adapts its drone tactics as fast as the models learn them.
Source: Kristóf Nagy, “Ukraine und NATO entwickeln gemeinsame KI-Lösungen zur Drohnenabwehr und Feuerunterstützung,” hartpunkt, October 16, 2026.