Picture a Ukrainian command center: screens show drones moving into a contested battlespace. Their pictures are relayed back to military operators, but they are also acting in concert through linked communications. Software helps determine which drone is best positioned to continue the mission based on its location, payload, and chances of survival.
The operator assesses the information collected by the drone flight, confirms the target, and makes the final decision on whether to engage. Artificial intelligence accelerates the process, but the human remains in control of the critical decision.
This is what coordinated autonomy means in practice. It does not remove people from warfare. It allows one operator to supervise multiple systems that can navigate, exchange information, and divide tasks even when satellite systems like GPS are jammed, or communications are disrupted. Without this, every drone must be managed separately, decisions take longer, and military units cannot operate at the speed and scale that modern warfare requires.
Ukraine is already developing this model in combat. NATO allies, on the other hand, possess advanced drones, sensors, and AI-enabled targeting platforms, but have not yet incorporated coordinated autonomy consistently into their doctrine, procurement, and training. NATO cannot treat autonomous systems as technology for a distant future. These capabilities are already reshaping warfare, and the alliance must adapt accordingly.
The defense industry often treats autonomy as a self-contained capability: a better algorithm, a faster processor, a smarter drone. In practice, it is none of those things in isolation. Autonomy is one part of a wider operational system that includes intelligence collection, data verification, mission planning, communications, human authority, and accountability after a mission. If one of these elements fails, the sophistication of the drone itself may not matter.
A system that navigates perfectly but strikes the wrong target is not a technical success – it is an operational failure. That is why the quality of the intelligence provided before a mission is so important. Information must be checked, accurately located, updated, and assessed for reliability before assets are committed. Once drones enter contested airspace and communications begin to fail, there may be little opportunity to correct a weak intelligence picture. Autonomy does not compensate for unreliable information; it can accelerate the consequences.
Since April 2024, Swarmer’s software has supported more than 100,000 combat missions in Ukraine, many conducted under persistent electronic warfare and in GNSS-denied environments. This experience shows that autonomy is not simply about making an individual drone smarter. It is about connecting multiple systems, reliable intelligence, and a human operator so that they can function as one operational unit.
Ukraine’s forces have developed practical thresholds for deciding what information is reliable enough to support a mission. However, many of these practices are unique to individual units, systems or mission types. NATO’s task is to study those lessons and turn them into common standards that can be applied across different countries, platforms, and suppliers.
This is a broader industry effort, not the work of one company. Swarmer is one of several providers developing software for coordinated autonomous systems, alongside platforms such as Auterion’s Nemyx and Helsing’s Altra. Because we develop these systems, we understand both what they can do and what they require to operate safely and effectively. The partnership between Swarmer and Molfar Intelligence provides one example of how the intelligence-autonomy connection can work. Molfar collects, cross-references, and structures open-source information, including geolocated imagery, satellite observations, and battlefield reporting. Swarmer can use verified datasets to improve object recognition, situational awareness and the reliability of autonomous systems. The point is not that NATO should adopt one commercial solution. It is that the connection between verified intelligence and autonomous platforms should become a standard requirement across the alliance.
For NATO to get autonomous operations right, there are three things that must happen now.
First, the alliance needs a definition of decision-grade intelligence for autonomous systems. In plain terms, it must decide how accurate, recent, and well-verified information must be before it can support an operational decision. The required level will differ depending on whether a system is conducting surveillance, navigating independently, or supporting a strike.
Second, procurement must treat the intelligence pipeline as integral to autonomous platform acquisition — not as a separate line item that someone else will figure out. When NATO buys an autonomous platform, it should also specify where its data comes from, how it is checked, how systems exchange it, and how human operators review it. Buying a drone swarm without specifying the data architecture is like buying a fighter jet without specifying the radar.
Third, the alliance should formalize the operational lessons that Ukraine is generating daily. Ukrainian operators, intelligence specialists and technology developers should be directly involved in shaping NATO testing, training and procurement. The alliance must distinguish between systems demonstrated in controlled conditions and capabilities proven under electronic warfare, disrupted communications, and the pressure of real combat.
NATO is not starting from zero. It has advanced technology, growing investment and a much greater willingness to learn from Ukraine than it did several years ago. What it lacks is time. Drone technology and battlefield practices are developing continuously, and future conflicts will not wait for NATO’s doctrine and procurement cycles to catch up.
The lesson from Ukraine is straightforward: autonomous systems cannot be purchased as stand-alone machines. They must be built around reliable intelligence, clear human authority, and common operational rules. NATO should establish those foundations now – before it is forced to learn the same lessons under fire.
Alex Fink is President and US CEO of Swarmer
James Westlake is Director of Molfar Intelligence’s UK Office
Europe’s Edge is CEPA’s online journal covering critical topics on the foreign policy docket across Europe and North America. All opinions expressed on Europe’s Edge are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
