The AI opportunity in LGR

Building AI-ready foundations in the new authority

Ben Court, Head of AI and Analytics, Civica

The councils that will extract real value from AI in the next five years are not the ones that buy AI tools, they're the ones that have begun to build AI-ready foundations. For those going through local government reorganisation (LGR) and already redesigning operating models, data estates and governance, there is a rare window now open to take a giant leap forward in setting up their new authority with core data and technology principles for AI enablement.

For chief executives and DDaT leaders, the challenge is balancing the demands of LGR today with the AI ambitions that will shape the next authority tomorrow. This article is designed to help you make the right moves now that will contribute towards achieving longer term AI goals in unison with delivering the critical steps of successful LGR integration.

LGR as a fork in the road

Before we start, however, it's worth putting the current state of play regarding AI adoption in most local authorities into context. Figures from Civica's recent research into Public Sector Financial Resilience, which surveyed senior professionals in local government and across the public sector, uncover a reality where councils are most commonly grappling with serious infrastructural blockers to AI enablement.

Only 30% said they have confidence in their data quality and governance, which is the bedrock of any AI deployment. Nearly two thirds (64%) said that they are only partially ready or less to scale digital solutions. And yet 11% cited employing AI as a top five priority in the next five years.

That gap is a story in itself. Ambition is outpacing infrastructure by a wide margin, and the reason is structural rather than a lack of will. Most authorities are trying to bolt AI ambitions onto data estates that were never designed to support them. But LGR is precisely the moment when those estates are being reviewed, merged and rebuilt.

This matters because LGR will either multiply existing data problems or create the conditions to fix them. If councils simply combine legacy systems, they risk carrying duplicated records, fragmented lineage and poor visibility into the new authority. If they use LGR to establish shared data principles, ownership and standards, on the other hand, they will be in a great position to create the foundations for safer, more effective AI.

Getting that sequencing right starts now.

Phase 1: Preparing for operational readiness

There is still a long road towards vesting day. This phase is about sequencing, not ambition. It's the no-regret actions that reduce operational risk before day one, regardless of how far AI ambitions eventually go. Skipping this stage doesn't just delay AI enablement later, it means the new authority could inherit the data problems of each of its predecessors simultaneously, and even exacerbate them.

To avoid this layer of complexity later:

• Map all data estates across the predecessor authorities

• Identify duplication, legacy formats and gaps

• Define data ownership and governance principles for the new authority

• Assess AI readiness score per service line.

Get that mapping right, and the new authority arrives at vesting day with somewhere to stand.

Phase 2: Operating as a safe and legal authority

Once the new authority is live, the priority shifts to proving financial grip, statutory assurance and clear accountability. AI must earn its place alongside those priorities, not compete with them. The authorities that get this right treat AI piloting as a byproduct of doing data consolidation properly, not as a separate workstream competing for the same scarce capacity.

In practice, this means you should:

• Implement unified data platform architecture

• Establish master data management for residents, properties and assets, as well as for internal ontologies among different teams e.g. account codes or cost centres in finance

• Create a single, trusted view of residents, properties, assets and services so that leaders can make decisions based on consistent information across the new authority

• Deploy data quality tooling, lineage tracking and a data catalogue and set baseline metrics

• Pilot AI use cases with clean, integrated data in lower-risk service areas and set clear, measurable outcomes to assess.

These steps create the trusted data environment needed to support more ambitious AI use in future.

Phase 3: Being future ready

With the foundational work done, the task becomes building a modern, high-performing council around citizens and place. This is where the investment in phases one and two pays off. Scaling AI becomes a natural extension of data work already completed, rather than a fresh transformation programme competing for attention against everything else the new authority is trying to prove.

This is where the effort shifts:

• Expand AI use cases across service lines with proven data foundations

• Establish AI governance and ownership within the new authority's DDaT function

• Build continuous feedback and improvement loops to account for citizen experience of the AI’s behaviour: AI performance, efficacy measures, adoption, outcomes

• Measure changes in efficiency, service quality and resident outcomes as part of LGR business case realisation.

The result will be a more connected, resilient authority that can use data and AI to improve outcomes for residents whilst maintaining the flexibility to respond to future challenges.

The impact of AI in the new authority

With these foundations in place, councils can begin applying AI in practical ways. This may include reducing administrative burden on staff, improving visibility of demand across services, supporting earlier intervention and helping leaders make more informed decisions based on trusted data.

Start with human-led, lower risk, high volume / high burden use cases and only move into more complex areas when governance and data are mature enough. This could be things like invoice ingestion, triaging workloads or summarising notes. In fact, this is where we can play to AI’s greatest strengths.

Deploying AI in areas such as revenues, housing and social care means making consequential decisions about people's lives. These decisions require explainability, clear accountability and strong governance from the outset. The councils that get this right will not just be AI-enabled; they will be AI-trusted.

Final thought

Local government leaders – from the executive team setting direction to the DDaT function delivering it – face a choice that will define the new authority for a decade: treat LGR as a systems migration exercise, or treat it as a data evolution opportunity. The former gets you to vesting day. The latter gets you to financial resilience and AI enablement.