Artificial intelligence (AI) is no longer a future ambition for the public sector. Technology partners are integrating AI directly into core enterprise platforms, moving it beyond experimentation and into day-to-day operations.
The direction from government is clear. The Artificial Intelligence Playbook for the UK Government sets out ten principles for safe, effective AI use across the public sector, including human oversight, transparency, skills and governance. AI should be adopted where it delivers public value, but only with accountability built in from the start.
Whilst the ambition is clear, delivery is a challenge. The National Audit Office reports that 70% of government bodies are piloting or planning AI use cases, yet most remain at pilot stage rather than scaled, operational impact. New research from Public First reveals why: more than half of UK civil servants (54%) have received no AI training, and just 39% describe AI as empowering. This is not a workforce that has rejected AI, it is a workforce that has been left to figure it out on its own.
The infrastructure of confidence matters more than the infrastructure of compute. This is where the conversation must shift.
From pilots to real impact
Despite strong national ambition, much of the public sector remains in the early stages of adoption. Parliamentary scrutiny has reinforced the message that ageing IT, poor data quality and a lack of transparency in algorithmic decision-making are putting AI adoption at risk. Across local government, higher education and blue-light services, the same pattern repeats – successful adoption depends less on the algorithms themselves and more on fixing long-standing issues with data, digital governance and the people who will ultimately use these tools every day.
Ultimately, AI success needs to be built on strong foundations, not fast implementation.
AI as a capacity multiplier, not a workforce replacement
Across the public sector, the gap between what is needed and what can be delivered is widening, and AI is increasingly being asked to help close it. AI should not be seen as a workforce replacement tool, but as a capacity multiplier.
Its greatest value lies in reducing administrative burden – automating repetitive tasks, improving forecasting and freeing up time for the work that matters most. Research from the Alan Turing Institute found that the biggest gains from AI come from taking on repetitive admin work, freeing people up for the judgement-based decisions only they can make.
This matters because of what it gives back to staff. Time to spend with a vulnerable resident rather than on a form. Time for an experienced caseworker to mentor a junior colleague. Time for a teacher to focus on the students in the classroom, or for an officer to be visible in their community. The goal is not fewer people. It is the better use of people.
The risk of focusing only on technology
Technology partners are moving fast. Oracle, for example, is now embedding AI directly into the core enterprise platforms many public sector organisations already rely on, with strong emphasis on data sovereignty and governance. Other major vendors are doing the same. The technology question, in many areas, is closer to being solved.
But step back, and a familiar gap appears in how AI is talked about. The conversation is overwhelmingly technical – model performance, governance frameworks, sovereign deployment, integration architecture. What is missing is meaningful focus on the people who will use these tools, and whose confidence will determine whether they succeed.
This is why so many public sector AI programmes stall. A social worker hesitates because no one has told them whether they are allowed to. An officer cannot stand behind an AI-prioritised case in court. A frontline team avoids it because they do not trust the data. If you treat AI as a purely technical upgrade, and value will quietly leak out.
What people-led AI adoption looks like
The organisations getting AI right are the ones treating it as a people change programme that happens to involve technology, not the other way around. In practice, that tends to look like five things:
- Clear permission from leaders: Staff need to know AI is encouraged, not just tolerated. Public First research shows that where leadership backing is clear, 91% of staff feel confident using AI, compared with just 45% where it is not.
- Role-specific training, not generic awareness: The same research shows that 75% of UK civil servants who have been trained find AI easy to use. The challenge is reach, not appetite.
- Visible safe spaces to experiment: Sandboxes, pilot teams and safe harbour rules let people try AI on low-risk tasks, building confidence through use.
- Peer-led learning, not just top-down rollout: Confidence spreads fastest when colleagues share what’s working; organisations that recognise and showcase teams using AI well move faster than those relying on training modules alone.
- Honest conversation with citizens: Public confidence is as important as workforce confidence. Being open about where AI is being used, and where human judgement still leads, is what protects long-term trust.
Human side of adoption
Frontline staff are often positive about AI’s potential, but everyday adoption still lags behind strategic ambition. The gap is rarely technical and more cultural.
Staff need to know what AI is doing on their behalf, where their judgement still matters most, and how to challenge an output that doesn’t look right. Citizens need to know that decisions affecting their lives remain fair, transparent and accountable. According to Appian, only 39% of UK citizens trust central government to use AI responsibly – a reminder that public confidence cannot be assumed. Leaders, in turn, need to understand that closing this trust gap is part of their job, not a side effect of getting the technology right.
This is what makes AI adoption a people change programme before it is a technology programme.
A call for balanced leadership
The technology is ready. Secure, enterprise-grade AI is increasingly available across the systems public sector organisations already use. The challenge for leaders is to match technical ambition with responsible adoption, strong change management and ethical governance.
AI can help the public sector do more with less. But only if it is implemented with people, not imposed on them.
Get in touch
We support public sector organisations through change and digital transformation that’s sustainable for employees, citizens and leaders. To talk to us about how we can help, get in touch today.