18 Nov 2025

From data lakes to decision engines: operationalising defence AI through integration by design

Guest blog by Balaji Anbil, Director - Applied Innovations at Tenacium DC Ltd #DefTechWeek2025

Balaji Anbil

Director - Applied Innovations, Tenacium DC Ltd

For defence leaders at a glance

Adopt architectural discipline: Treat architecture as the next frontier of Defence transformation, guiding every AI and data decision.
Integrate by design: Build systems that connect legacy and modern capabilities, enabling humans and AI to share context, insight and accountability.
Command the AI cycle: Move beyond hype to master the full continuum from Context to Consequence. Understanding the weights, acting on the responses, and interpreting the decision trees will define future national advantage.

Every decade, Defence faces a moment where technology stops being a tool and becomes a terrain. We are in one of those moments now.

After the AI hype settles, the nations that master the full AI cycle, from Context to Consequence, will define the next century of technological and strategic advantage. Power will no longer rest with those who build the largest models, but with those who can interpret context, act with precision, and understand consequence. The true edge will belong to those who architect AI responsibly and integrate it coherently across national systems.

The question is not whether data and AI can transform the Defence Enterprise. The question is whether we can build the right architecture to make that transformation coherent, safe and explainable.

The UK’s Defence Technology strategy sets a bold direction. Yet ambition alone will not make digital integration real. The next frontier is architectural discipline. To operationalise AI safely, Defence needs to move from collecting information to interpreting it, from experimental pilots to trusted, explainable systems that serve people and missions.

The future battlespace

The battlespace has now become a dynamic, real-time, digital theatre of command and control. It is a living web of sensors, autonomous platforms and human decisions, stretching from space to subsea. Within the next decade, drones, robotic units and distributed AI agents will act alongside the military in the battlespace.

Yet beneath this transformation lie persistent data challenges that prevent Defence from realising its full digital advantage:

Fragmented data chains: Sensors, platforms and command systems collect but rarely correlate information, leaving gaps in shared context.
Excess without meaning: Petabytes of unstructured data obscure the few insights that matter for timely and decisive action.
Reactive governance: Oversight often follows the data rather than shaping it.
Pilot fatigue: Experiments in AI and autonomy struggle to move seamlessly from proof of concept to sustained capability.
Human overload: Skilled staff spend more time reconciling data than interpreting it.

As Defence architectures evolve, intelligence will no longer live solely in central clouds. Small, task-specific language models and edge-operated devices will process data close to the source, reducing latency and dependency on connectivity. Federation between systems will become essential, allowing secure collaboration across allies and domains without exposing raw data.

These principles must sit within a secure-by-design framework so that every model, sensor and algorithm is trustworthy from inception rather than patched for compliance later.

Most organisations have tried to fit AI into Defence data environments. Few have successfully architected data for AI. Our work in building JEDAI (Just Enough Data Analytics and Insights) our Gen-Ai situational awareness platform has shown that with the right architectural discipline, this can change. Turning fragmented data into a coherent, explainable decision system.

Integration by Design is therefore not about technology choice. It is about intent. When systems are designed to share context, explain reasoning and preserve human oversight, AI becomes an ally rather than a mystery.

Lessons from architectural practice

At Tenacium DC, we learned early that the answer to data overload is not more analytics but better structure. Architectures that refine before they automate can turn noise into meaning.

The principle is simple: collect everything, but model only what matters. When data is organised around relationships between people, systems and missions, information becomes provable rather than probabilistic.

This thinking shaped our approach to the AI data refinery model. Instead of replacing or migrating legacy systems, we created an architecture that sits alongside them, refining and enriching their outputs into shared, explainable context. It is built to coexist, not overwrite. Migration at the scale Defence requires is not practical in the near to mid term, and so coexistence must become a design principle, not a compromise.

What differentiates this approach is that it begins with architectural empathy rather than control. It treats existing systems as valuable sources of truth, not obsolete infrastructure. By layering explainability and data lineage across them, we can achieve integration without upheaval.

At the same time, we must remember that AI is not an intelligence in itself. It is a mathematical construct built on functions, weightings and probabilities. It cannot intuit or empathise. Preparing the Defence workforce to understand this distinction is as important as deploying the models themselves. A future-ready force is one that can interpret what AI produces, not simply trust it.

Standards as strategic infrastructure

To operationalise Defence AI, the Ministry of Defence must do more than procure technology. It must publish architectural standards that suppliers can follow. Just as the Digital Technology Reference Model set the foundation for zero-trust and data-centric security, an equivalent reference model for AI-ready architectures is now essential.

Future Defence standards must reflect these realities. Guidance on small LLM deployment, data federation and edge security should accompany principles of interoperability and explainability to keep innovation safe, scalable and sovereign.

Clear guidance on data lineage, interoperability, and explainability will enable SMEs, primes and academia to innovate within a shared framework. Without these standards, every new AI pilot risks becoming another isolated experiment. With them, Defence can build an ecosystem where innovation compounds rather than fragments.

Architectural standards are the grammar of integration. They define how trust travels through systems, how accountability is maintained, and how decisions can be traced even when machines are involved.

From theory to trusted practice

Operationalising AI is not about building smarter algorithms. It is about creating environments where human expertise and machine logic can reason together. That means every AI deployment must have three qualities: explainability, interoperability and human oversight. Systems must be able to show not only what they decided, but why.

Tenacium’s work with JEDAI demonstrates that these qualities can coexist with speed and agility. When integration and trust are designed in from the start, AI becomes more predictable, evidence becomes reusable and operational tempo increases without sacrificing safety.

For Defence, this is the path from data accumulation to decision advantage. It turns digital infrastructure into a living intelligence system that learns responsibly.

A shared call to action

The future of Defence will be built by an ecosystem that speaks a common architectural language, guided by intent for integration and explainability. The Ministry of Defence can lead that conversation through open standards. The next decade of Defence technology will not be defined by who has the most data, but by who interprets it best.

Defence leaders should expect their supply ecosystem to deliver more than software. They should demand clarity, traceability and explainable collaboration. The systems they procure must:

Provide a true operational picture continuously and in near real time
Build trust through transparent data lineage and provenance
Accelerate human-machine collaboration grounded in shared context

Once the current wave of AI hype fades, the measure of national strength will not be who has the most data or the largest models, but who can turn understanding into action and action into informed decision. The nations that master this full AI cycle, from context to consequence, will lead the next century of technology-driven advantage. Integration by Design is how the UK Defence Enterprise can get there first.

#DefTechWeek2025-Tenacium DC Ltd-blog-pic1.png

Operationalising DefTech campaign week 2025 #DefTechWeek2025

techUK has launched its Operationalising DefTech 2025 campaign week, highlighting how emerging technologies can strengthen defence capability and accelerate innovation across the sector. The campaign showcases member insights, practical use cases, and opportunities to collaborate on the future of defence technology.

Read them here

Operationalising DefTech: Defence campaign week launch

Launching techUK’s Defence Campaign Week, Operationalising DefTech, techUK is pleased to welcome John Cunningham, the Director of UK Defence Innovation within the Ministry of Defence to brief members on MOD’s evolving approach to new technologies, alongside the co-founder of Anduril Industries Christian Brose, and Sara Sharkey from Prolinx.

Click here to register