Think AI for Government: Key Takeaways

Momentum around AI in the public sector is accelerating, with the government moving from small-scale experiments to deploying real-world solutions that promise to transform how public services are delivered, designed and experienced.

But what will that look like? That was the topic of the day at the Think AI for Government event in London this week. Alongside technology suppliers, representatives from a host of government organisations debated AI’s role, and potential applications, across the public sector.

“With 25 brilliant speakers across nine insightful sessions, the Think AI for Government conference has managed to raise the bar following on from the launch event that we ran earlier in the year,” said Think Digital Partners founder, Matt Stanley. “There was so much positivity in the room around the future of artificial intelligence across the public sector. With effective governance and outstanding innovations, this is a very exciting place to be right now.”

The opening government keynote detailed the UK government’s accelerating efforts to integrate AI across the public sector, supported by initiatives like the £42 million AI Frontiers Fund and projects led by the Government Digital Service (GDS).

It highlighted how GDS has rapidly evolved from developing basic productivity tools such as transcription and data analysis aids to driving system-wide transformation in key areas like health, education, and planning. A focus was Extract, an AI tool that digitises complex planning documents to speed up approval processes and support housing and economic growth.

But while some departments, like the Ministry of Justice and HMRC, are leading in AI adoption, progress across government remains uneven. To address this, GDS is building AI capacity, promoting agile experimentation, and learning from private-sector lessons about aligning technology with real-world problems.

From Reactive Services to Agentic Systems

Another session explored how AI is reshaping public services by shifting from reactive delivery to more proactive, agentic systems that can anticipate citizens’ needs, personalise support, and improve efficiency.

Speakers from government, industry and the tech industry discussed how AI is moving beyond experimental pilots toward real-world implementation. They highlighted examples of AI applications in justice, welfare, and administrative processes, noting that AI should serve as an enabler – not a replacement – for human-centered service design.

“We need to start looking at re-imagining the experiences we want to deliver for citizens and work from there; production readiness and all the technical pieces will follow,” noted Deepak Shukla, public sector data & AI strategy Lead at AWS.

Partnerships Powering Progress

Elsewhere, government-industry partnerships were put under the spotlight, with an expert panel determining that such relationships are essential to scaling AI responsibly and efficiently across public services.

While SMEs bring agility and deep expertise, they face barriers from rigid procurement and risk aversion. However, with new powers under the UK Procurement Act, departments can now adopt more flexible, demo-based and dynamic procurement to engage innovators faster.

At the same time, building trust through transparency, explainability and open-source collaboration are viewed as non-negotiable, alongside smarter decisions about when to build in-house versus buy from the market.

“Trust is non-negotiable in national security… explainability [is] mandated… Government people need to understand [so they’re] really making a robust decision,” said Chad Bond, director of strategy and innovation at Zaizi.

Drawing lessons from countries like Estonia and Ukraine, the panel argued that successful partnerships require shared goals, technical fluency on both sides, and a co-design approach where government acts not just as a buyer, but as a collaborative partner in delivering public value.

Building an Ethical, Inclusive Future

So what does the future of AI in government hold? The Future of AI in Government panel painted a picture of a public sector on the edge of transformation. Speakers agreed that this moment feels different, with leadership attention, investment and ambition are finally aligning.

“AI transformation only succeeds when organisations accept it, adopt it and adapt it,” said JP Bhamu, director of data and AI at the NHS Business Services Authority.

He pointed to “something fundamental” changing at the highest levels of government, citing the new AI Opportunity Action Plan and the creation of AI Growth Labs signals of serious intent. But optimism was tempered by realism.

“Public sector organisations are far too complex,” he warned, arguing that data silos, legacy systems, and rigid structures could still slow progress. The challenge, he said, is to make AI adoption both fast and safe – a cultural as well as technical shift.

For others, the heart of the discussion lay in how to make this AI-driven future inclusive, ethical, and human-centred. “The ‘I’ in AI has to be the starting line and cannot be the finish line,” said founder of Impact Circle, Kanika Joshi, who called for keeping “people from vulnerable communities first in decision-making.”