A Practical Step Change in Decision Support
Generative AI is recognised as a force multiplier. It offers a new class of capability for Defence decision-making - not by replacing human expertise, but by improving how information is interpreted, how options are generated, and how decisions are made. Unlike traditional AI, which is often rules-based and task-specific, generative and agentic AI systems are designed to handle ambiguity, generate new content, and respond dynamically to changing contexts.
This makes them particularly well-suited to supporting military decision-makers, where speed, situational awareness, and adaptability are critical. [Learn about starting AI here]
Where Generative AI Delivers Improvements
By integrating generative AI into the human-machine interface and ensuring training and governance are built in from the start, Defence can achieve a meaningful advantage at key points in the observe, orient, decide, and act (OODA) loop. We see three areas where generative AI meaningfully changes how decisions are made in Defence, including within Joint All-Domain Command and Control (JADC2) frameworks and multi-domain operations (MDO):
- Smarter Observation
Generative AI can synthesise large volumes of diverse data sourced from structured military systems to unstructured, open-source inputs; and help to create coherent summaries that support faster understanding. This differs from traditional automation by enabling richer context, interpretation, and reasoning, not just data processing. It improves how humans comprehend complex environments and accelerates the transition from raw data to insight. - Faster, Cross-Domain Orientation
Generative models are pre-trained and flexible - they’re not locked to a single use case or dataset. This allows them to act as connective tissue between domains and platforms. They help Defence integrate cross-domain intelligence and operational information from land, sea, air, space and cyber, improving sense-making and coherence across silos. Orientation becomes not just faster, but more complete, moving from analysis of the known to anticipation of the emerging. - Scenario Generation and Decision Support
Generative AI's unique functionalities represent a game-changing capability for boosting human team collaboration. It increases the speed at which teams can take synthesises data and generate ‘what-if’ scenarios, such as alternative courses of action; and the speed at which they can be shared, reviewed and assessed, which subsequently improves the planning cycles and leads to more robust outcomes.
All of this this leads to faster planning cycles, better collaboration, and more informed decisions - while keeping humans firmly in control.
Trust as a Feature, not a Constraint
While the potential is clear, realising value from AI depends on trust. Many commercial AI initiatives have failed; not because the AI systems were weak, but because the governance was weak. Defence must take a different path. [Lessons about AI in the finance sector here]
Like the ‘brakes on a car’ - the faster it’s meant to go, the stronger the brakes need to be – good governance is vital to build AI systems at speed, and with confidence. IBM has long advocated for AI that are transparent, auditable, and secure. In high-consequence environments like Defence, this is not optional, it is foundational. Strong governance does not slow progress; it enables it.
As a long-standing partner to Defence and a first mover in generative AI deployment, IBM continues to prioritise trust, accountability, and mission alignment in every solution. [Learn about AI Governance here]
Rather than treating AI as a one-off project or tool, Defence can treat it as a sustained capability - designed with humans in the loop, governed responsibly, and applied where it makes the most operational sense.