The UK government’s up-to-date AI planning prototype built using Gemini aims to halve the time it takes to process homeowner applications
Governments around the world are exploring how artificial intelligence can deliver better public services faster. The UK is working to build 1.5 million up-to-date homes by 2029, but work is often slowed down by local planning authorities due to high levels of paperwork and backlogs. To aid build Britain, we are working with the UK Government to dramatically reduce processing times for household planning applications. Our aim is to aid officials reduce the time it takes to decide on applications by 50%, freeing up planners’ time to build more homes. We’re excited to see how our national AI partnerships, which aim to support a rethink of public services to create more resilient societies, can aid the UK grow faster.
Co-creating a up-to-date AI planning tool enabling faster decision-making
After successful launch Extract – a tool created with Gemini by the UK Government’s AI Incubator (i.AI) – to aid local governments transform elderly planning documents into clear, digital data – we are working together to develop a up-to-date, sophisticated planning prototype for local governments.
Google DeepMind is working with the UK Government, Google Cloud, the department and local planning authorities in Barnet, Dorset and Camden to co-develop an AI-powered prototype that acts as a highly skilled assistant to planning officers tackling challenging data extraction and case analysis operations. Close cooperation will enable us to build an optimized tool, tailored to the unique needs and everyday challenges that planners face on a daily basis. Following early trials in Barnet, Camden and Dorset, the Government plans to make the up-to-date AI planning tool available to all councils across the country from 2027.
In a typical up-to-date planning application, officers spend hours comparing policy documents, historical files and PDFs. This manual process creates a grave bottleneck. This is particularly challenging given that household applications make up almost 70% of planning applications each year. By reducing time spent on plain matters such as loft conversions or extensions, the prototype being tested in Barnet, Camden and Dorset could aid planning professionals focus more on complicated public benefit applications.
The AI scheduling tool will streamline routine tasks using AI by:
- Data consolidation: Pre-process backlogs, highlight data gaps and extract key site information so planners can view everything on one screen.
- Identifying local policies: Highlighting relevant national and local policies, pre-assess compliance and provide correct citations for review by the official.
- Review summary: Reviewing individual consultation letters to identify key concerns or precedents.
- Preparation of assessments: Creation of the first draft of the final report, containing justification and proposed conditions.
Most importantly, the planning officer has full control as the final decision maker. They review every line generated by the tool, edit the justification, and retain the authority to approve or reject the request. To ensure accountability, the prototype records its work at every step, creating a clear chain of thought and a solid audit trail for every decision to support the planning professional.