How AI can fix planning

Planning in the UK is slow, complex and opaque. The government has outlined various plans to address this, including overhauling housing targets and revising the national planning policy framework, as well as most recently publishing its 10-year Infrastructure Strategy. The aim is to introduce a more integrated, spatial approach to planning and public services.

But relatively less noticed was the Prime Minister's announcement at London Tech Week about Extract, an AI project that transforms analogue documents into digital planning data. Is this the start of AI eating planning? Or will it just be another fad of digital transformation broken on the wheel of professions? At our latest Digital Forum, we had the pleasure of hosting a conversation on precisely this question. Three expert guests joined me: 

• Duncan Brown, Head of Software Engineering at i.AI

• Paul Maltby, Director of Public Services at Faculty AI

• Nissa Shahid, Chartered Planner and Digital Lead at Arup

Together, we explored the opportunities and risks of using AI in planning, and the role of digital innovation in shaping its future.

From the ‘reign of experts’  to empowered citizens

Planning has been dominated by “the reign of experts”; professionals with access to specialist knowledge, complex rules and highly technical processes. But as AI tools become more accessible and powerful, that expertise can be distributed more widely. They could create a system where everyone, from professionals and citizens, can engage more meaningfully in decision-making.

AI’s ability to process large amounts of complex data and language, and translate it into accessible insights, could help both make the system work quicker and demystify the planning system for non-professionals. Not to displace expertise, but to distribute it, giving people the ability to develop more relatively expert views. 

The focus now is on making processes more efficient and removing the drudgery of administration. Extract uses AI to convert dense legal documents into structured, digital data, making it dramatically faster and easier to process planning records. These tools are laying stronger digital foundations to enhance decision-making.

How AI can make planning more transparent and human

Technology alone won’t transform the planning system. Success depends on how we design and deploy it, and whether we do so with the end user in mind.

One panellist raised the risk of an AI arms race. Too many planning documents are written in dense, inaccessible language. How do we manage the risk of AI generating long planning documents that then others use AI to summarise 

Instead, the opportunity for AI lies in redesigning how information is communicated entirely, putting clarity and context first.

When designed properly, AI can also help make the planning system more transparent by sorting through and linking the myriad reams of different planning documents and data. AI  can make it easier to trace decisions, see how evidence is used, and hold the system to account. But that only works when the design is intentional, and we step back and look at the problem we need to solve first and focus on people as well as processes.

Embracing experimentation to push progress

When it comes to introducing AI into planning, the panel stressed we won’t get it right by over-planning or waiting for perfect solutions. We need to move beyond the usual discussions of buy or build when there’s a third option, ‘mess around’. Local Authorities need to feel confident to experiment, testing tools in live settings, learning quickly, and iterating based on what works.

AI adoption should feel exploratory, not overwhelming. Low-commitment experimentation is critical for both innovation and building internal confidence and capability.

This approach also helps bridge the traditional gap between digital teams and built environment professionals. Many of the best breakthroughs in planning AI are coming from non-specialists who’ve been empowered to punch through that barrier and build something useful. Experimentation creates space for creativity, collaboration, and better tools.

Automating the drudge work and freeing planners to plan

It was agreed that AI's most immediate value lies not in replacing planners but in automating repetitive and time-consuming administrative tasks. This would free planners for the truly critical, creative work where professional judgement is required.

AI can transform access to planning data currently locked in PDFs or lacking in structure and standard, tackling perennial and really boring problems, like locating correct documents on council websites. 

A smarter, fairer future for planning

The conversation at this Digital Forum demonstrated that AI will definitely have a significant impact on planning but what impact and where is still up for grabs. We heard how AI could help create a better planning system that’s faster, fairer, and built around the needs of both professionals and the public.

This won’t happen overnight. But with thoughtful design, shared learning, and a willingness to try new things, we can build planning systems that serve people and places far more effectively than today.

We’re committed to rethinking how technology can serve people, places and the planet. Want to be part of the next discussion? Register your interest and keep an eye out for our next Digital Forum.