If there was one message we took away from the recent Global AI Cities event in Manchester, it was this: there’s a huge difference between a council that is AI-able and one that is AI-ready.
Today, so many local authorities (and those within the wider public sector) already have access to AI tools, meaning it’s relatively straightforward to be AI-able. AI readiness, however, is something entirely different, a process that requires organisational trust, strong governance, quality data, engaged staff, and public confidence. Without these foundations, even the most impressive technology risks delivering very little value.
As we’ll explore, the gap between AI-able and AI-ready might seem small, but there is a difference between the two concepts. With this in mind, it’s time to understand how organisations can make that leap from the former to the later, bridging a space to guarantee their future success.
Trust is the real starting point
For local government — and the public sector at-large — AI adoption is not simply about technology, but about trust.
Residents need confidence that their council is using AI responsibly, transparently, and in ways that improve services. Trust is not created through policy documents alone; it is built through engagement, participation, and consistent delivery of services.
The public’s relationship with AI is not abstract; rather, it is influenced by every interaction they have with public services and every decision they see being made about data. Because of this, councils cannot assume trust exists. Instead, they must actively earn it.
This creates a participation challenge. If AI systems are shaping how services are delivered, residents need opportunities to understand, question, and contribute to the conversation. The future of AI-enabled government will require collaboration between public agencies and the communities they serve.
Augmentation, not replacement
Another recurring theme in Manchester was the need to change the narrative around AI and the workforce.
Too often, conversations about AI focus on automation and cost reduction. While efficiency matters, local government and the greater public sector should be thinking much more broadly. AI’s greatest value is its ability to augment staff, helping them to spend less time on repetitive tasks and more time delivering meaningful outcomes for residents.
The most successful councils — those that are truly AI-ready — will take their people on an iterative journey of engagement, experimentation, training, feedback, and delivery. AI readiness is not achieved through a single deployment, but carefully developed through continuous learning, a process of hands-on experimentation that builds confidence and knowledge.
Government AI is not corporate AI
The event also highlighted an important distinction, one that is sometimes overlooked by those within the sector: government AI is fundamentally different from corporate AI.
Consider this: while private sector organisations often optimise for competitive advantage, revenue growth, or productivity gains, local government must balance efficiency with accountability, transparency, fairness, and public trust.
This distinction has important implications for governance and data management. After all, public sector and local government organisations cannot simply copy approaches from commercial environments; they need frameworks designed around public service principles and democratic accountability.
That is one reason why data sovereignty continues to top local government agendas. Councils need confidence in where data is stored, who can access it, and precisely how it is being used. For UK public sector organisations, these questions are becoming increasingly important, especially as AI capabilities expand.
Data is the foundation of success
If AI was one of the dominant topics in Manchester, it is safe to say that data was the topic of many conversations during the event.
Before organisations can expect meaningful outcomes from AI, they must ensure their data is fit for purpose. Leaders should ask themselves, “Does our data measure what we think it measures? Is it current, usable, and trustworthy? Has it been properly maintained?”
AI has many uses, but more than anything else, it has amplified the value of data as a strategic asset. In many cases, organisations are discovering that their biggest challenge is not adopting AI but understanding the quality and completeness of the information feeding it.
This also presents a security issue because — as the value of data increases — so does its attractiveness to hostile actors. AI strategies should therefore be designed with a zero-trust mindset, ensuring security, governance, and resilience are built in from the outset.
At the same time, innovations such as small language models, synthetic metadata, and domain-specific AI solutions are creating new opportunities to balance performance, privacy, and control.
Focusing on outcomes, not effort
Perhaps the most exciting shift discussed at the event was the move from effort-based operating models to outcome-based ones.
Historically, success has often been measured by activity (i.e., how many forms were processed, how many calls were answered, or how many hours were spent on a task, etc.), but AI creates the opportunity to think differently.
The real measure of success becomes the outcome delivered for residents. Was the issue resolved faster? Was the experience improved? Was staff time redirected towards higher-value work?
Successful AI adoption — the ability to be truly AI-ready — comes down to three factors: people, process, and proof. People need the skills and confidence to use AI effectively. Processes must support a “verify, then trust” approach to decision-making, and organisations need proof that AI is increasing operational velocity and improving outcomes.
Those that get this balance right will not simply be AI-able, but genuinely AI-ready — and that is where the real transformation begins.
