Ukraine Shows AI Changing How Coalitions Fight

Coalition warfare has always been messy. Different procedures, systems, standards, doctrines, and operational caveats complicate and slow coordination at every level of conflict. AI may finally cut through that dynamic to become the connective tissue that makes multinational forces more cohesive and coherent.

Marco Criscuolo, NATO’s acting deputy assistant secretary general and director for strategy and policy with the Cyber and Digital Transformation Division, likened managing NATO’s multi-domain operations to solving a three-pronged problem. On one axis lie the ‘five operational domains’, on another the ‘strategic, tactical, and operational levels’ and finally ‘the five classification domains, from NATO unclassified up to top secret’. The challenge of making real-time decisions is complicated not only by coordinating these domains, but also by the presence of multiple militaries. According to Criscuolo, it ‘is not possible’ to do so effectively without AI capable of quickly processing mass data points.

On the panel, Lyuba Shipovich, chief executive and co-founder of charity Dignitas Ukraine, raised the unique challenges of modern warfare in Ukraine, asserting that the physical dimensions of conflict are evolving. The heavy concentration of drones is creating a transparent battle space. As a result, Shipovich contends that old military standards, procedures and protocols may no longer apply, ‘everything has to be adjusted and has to be adjustable all the time’, demanding greater flexibility and speed—which is where AI solutions can make an impact. Adaptive medical protocols is a case in point in this evolving battlefield environment.

Another important factor in optimising AI for interoperability is the role of ensuring a human being in the loop, on the loop, outside of the loop. As Criscuolo explained:

It’s about also putting the developers closer to the end user with a very agile way of developing software and to move away from a hardware-based development and into a much more software-defined defence. Drones on the shelves, would be unusable the day after. So, we need to decouple hardware from software. It is also about how we innovate, long term versus short term. NATO has recently approved a rapid action plan, which will allow us to really benefit from early adoption, but also to support industrialisation, sustainment and cavalry developments.

In this regard, much has to do with how interoperability is incentivised—and this is where moving to machine-readable digital standards can be put into the field and made available to industry to undertake tests and evaluations that are iterative in operational environments.

The panellists noted that having effective software for weapons systems is a critically important part of warfare, with rapid updates facilitating the adaptability required by the modern battlefield. In this vein, AI has the potential to power the most adaptable forms of military software—capable of making real-time decisions in constantly changing circumstances. AI can also serve as a force multiplier for alliances by allowing smaller states to squeeze the absolute most out of their capability, particularly when facing larger states. Panellist Airis Rikveilis, national security advisor to the Latvian prime minister, stated that modern technologies can provide smaller states with ‘asymmetric advantages’, ultimately increasing the collective effectiveness of the whole coalition. AI may thus not only increase efficiency and interoperability, but also the effectiveness of each component military—significantly increasing the effectiveness of a joint force.

In an AI-powered world, humans’ role in warfare is driving change. On the podcast, Criscuolo outlined that as AI advances, the human role in multi-domain operations will move from ‘full control’ to the ‘orchestration of assets’ which could operate semi-autonomously. In this scenario, AI-powered autonomous drones may not require a pilot, but instead a human manager, to oversee actions and provide input when ambiguity arises—maximising efficiency and minimising errors. This is a significant shift in how warfare is conceptualised, and Criscuolo stressed the need for transparency in the adoption of AI given legal and ethical implications. Notably, adversaries may not feel similarly constrained by such limitations.

As the future of warfare plays out on the other side of the globe, Australia has much to gain from watching how the Euro-Atlantic adapts to new realities. As Australia pursues key capability initiatives, such as AUKUS, which focus on technological innovation and multi-national interoperability, AI could enhance efficiency, productivity and interoperability. Overcoming those traditional coalition constraints would improve cohesion and effective collaboration in the development and participation in activities from training to deployment.

The Indo-Pacific should draw on lessons from the Euro-Atlantic now, to shore up greater legitimacy and potency of future coalition operations.

Courtney Stewart is deputy director of ASPI’s Defence Strategy program and Carl Janz is a research intern in ASPI’s Defence Strategy program.

AI contributed no ideas to this article—Courtney Stewart and Carl Janz.

This article appeared originally at The Strategist (ASPI).