From concept to operational service: Thales, RAF Digital and NAD deliver AI milestone for Multi-Domain Mission Support System

• Europe
• United Kingdom
• Naval

The measure of AI in defence is not what it's capable of in a laboratory. It is what it can achieve in service.

A system built for the complexity of modern operations

MD MSS has been in continuous operational service since 1986. What began as a mission data system for the Nimrod Maritime Patrol Aircraft has evolved, across four decades of spiral development with the UK MOD, into the RAF's primary system for multi-domain mission support.

The current contract, awarded by DE&S (now NAD) in 2021 and extended in 2026, sustains and develops a system used across a broad range of platforms, including the F-35B Lightning II, Typhoon, Poseidon, Voyager and Chinook. MD MSS is also installed aboard HMS Queen Elizabeth and HMS Prince of Wales, supporting air operations by the Carrier Strike Group.

At its core, MD MSS processes, transmits and manipulates complex mission data, from multiple sources and locations, to deliver critical information at speed to aircrew and ground-based planners. It provides the situational awareness and communications interoperability that modern multi-domain operations require, assisting effective and rapid decision-making at both the operational and tactical level. It is this foundation that the latest programme milestone builds upon.

A step change, not an update

MD MSS has long incorporated analytical tools, including pattern-of-life visualisation and behavioural alerting, that help operators manage the volume of data flowing through the system. These capabilities, while battle-tested over years of operational service, were rule-based: meaning they responded to predefined parameters rather than learning from the data itself.

What has now entered service is of a different order. Using the Thales cortAIx Factory's Frugal Learning methodology - developed in collaboration with Faculty AI - Thales has deployed the first true machine learning capability within MD MSS.

This system trains on limited operational data to build an adaptive model of normal and abnormal behaviour. Initially applied to maritime track analysis, it monitors shipping patterns, filters routine activity, and identifies anomalies that would not otherwise be visible within the volume of data available to an operator. Critically, it does this using the kind of limited, real-world operational data that is available in practice, rather than the large curated datasets on which many other AI systems depend.

From concept to capability

This capability did not emerge from a standalone research programme - instead it grew out of a deliberate effort by the Thales UK cortAIx Lab to explore how machine learning could be applied across defence use cases to deliver genuine operational advantage from limited operational data. That initiative, termed AI DAMS - AI for Decision Advantage and Mission Support - was developed and tested in collaboration with Faculty AI through early prototypes to prove the concept. Thales then approached RAF Digital and DE&S with the proposal and, working collaboratively, built a more mature demonstrator using MD MSS's existing secure application hosting and data fabric capabilities.

User engagement, facilitated by the RAF and DE&S, refined the capability over successive iterations. Underpinning the programme's progress was the alignment of four conditions that too often remain unresolved in defence AI programmes: a credible technology solution, access to appropriate data, a clearly defined user need, and a viable route to exploitation. Without all four, these programmes typically stall. With all four in place, AI DAMS moved from initial concept to operational service.

The integration of AI within MD MSS represents another step ahead with enhancements to this RAF operational capability. Interrogating and interpreting complex mission data at pace will directly support more timely and informed decision-making in demanding environments. From a NAD Materiel perspective, this milestone demonstrates how cutting-edge digital technologies can be rapidly and effectively embedded into proven, in-service systems, reinforcing the value of sustained collaboration between MOD, RAF Digital and industry partners.

Alex Buckley - Team Leader NAD Materiel

This is a significant step forward for the RAF's use of AI in operational systems. By helping operators analyse complex mission data faster and focus on what matters most, this capability supports quicker, better-informed decision-making. It also marks an important milestone for MD MSS, showing how AI can be integrated into a proven operational capability in a practical and mission-relevant way.

Terry Makewell - RAF Chief Digital & Technology Officer

A framework for the future

Maritime track analysis is the first application of Frugal Learning within MD MSS, but it establishes a foundation for further development across additional track types and domains. The AI framework now embedded in the system can support expansion to flight tracks and other data sources as the programme evolves.

Beyond MD MSS, this deployment also represents something of wider significance: the first operational delivery of Mission focused AI from the Thales UK cortAIx Factory, the Group's dedicated AI accelerator. While cortAIx has broad reach within Thales globally, this marks its first delivered capability into UK military service - and, having established a repeatable, collaborative model for taking AI from research into operational deployment, it opens the door to further AI capability development across the UK Armed Forces.