Artificial Intelligence's Role in Planning and Construction: A Friend or a Threat?

The UK housing and planning sector is undergoing a digital revolution, with the integration of Artificial Intelligence (AI) aiming to modernize, speed up, and streamline processes. This transformation is set to deliver new homes and infrastructure more efficiently[1].

One of the key advancements is the use of AI tools that digitize planning documents, converting scanned or handwritten documents into high-quality digital data in under a minute[1]. This innovation is expected to help local councils process applications more swiftly, reducing backlogs. The Planning and Infrastructure Bill is incorporating this development to expedite decisions, support the construction of 1.5 million homes, and accelerate infrastructure delivery while maintaining environmental standards[1].

Beyond speeding approvals, AI-powered Property Technology (PropTech) solutions are being scaled to address challenges in planning and housing delivery. The UK government has initiated PropTech Innovation Challenges and funds to encourage local planning authorities’ adoption of digital tools[3]. AI is also being used by social landlords for automated workflows and chatbots to improve operational efficiency in managing housing services[4].

The benefits of AI implementation are numerous. Faster decision-making and streamlined planning processes are expected to reduce delays and backlogs[1]. Improved accuracy and data quality through digitization of historical planning data will also be achieved[1]. AI is expected to support ambitious housing targets and infrastructure goals, contributing to economic growth and energy security[1][3]. Enhanced public service delivery through better use of location-specific data is another potential advantage, applicable also to sectors like healthcare, education, and transport[1]. The growth of the PropTech sector, projected to expand significantly, will generate substantial economic revenue and investment[3].

However, concerns and challenges associated with AI in housing and planning exist. Transparency and accountability issues are a concern, especially when AI influences public procurement and planning approvals[2]. Legal and fairness considerations, such as ensuring AI does not introduce bias or reduce public trust in planning decisions, are also important[2]. The need for significant digital infrastructure and power, as AI tools and data centers require robust energy and planning support, is another challenge[5]. The complexity involved in integrating AI ethically and effectively across multiple public services without exacerbating social inequalities or reducing human oversight is a significant concern[2].

Despite these challenges, AI is poised to dramatically transform the UK's housing and planning sector, making it more data-driven, efficient, and responsive. However, it raises important questions about governance, infrastructure, and equitable implementation[1][2][3][4][5].

Considerations of design or the impact of a proposal on heritage assets are subjective, and AI could potentially be used to review minor planning applications such as householder applications, Certificates of Lawfulness, or conditions discharge. AI could also improve efficiency and decrease cost in the planning sector by generating and analyzing housing or employment projections, reviewing and categorizing site submissions, managing consultation, and auto-generating reports and analyses[6].

However, the tailoring of consultation to what a computer perceives as one's interests could lead to an artificial narrowing of options, or reinforcement of filter bubbles. AI has no intrinsic agency and no accountability, and its output must be evaluated by an accountable human[6]. Developers working across different authorities could benefit from the standardisation of validation requirements[6].

Planning relies on the exercise of judgement and the weighing up of the planning balance. Planning in the UK is not a 'tick-box' exercise and should not be[6]. Decisions must have some kind of democratic oversight to ensure public good is balanced against private interest[6]. One third of planning applications contain basic errors, and negotiating good planning outcomes will continue to require human actors to exercise nuance, common sense, creativity, and critical judgement[6].

The Alan Turing Institute has run pilots on automating application validation using machine learning[7]. Automation could enable the targeting of more specific groups on a Local Plan or application consultation, aiding public engagement[6].

In conclusion, AI has enormous potential to speed up the development process for applicants and authorities, and improve the housing and planning sector. However, it is crucial to address the challenges and concerns associated with its implementation to ensure a fair, transparent, and accountable process.

[1] Planning and Infrastructure Bill (2020) [2] House of Lords Select Committee on Artificial Intelligence (2018) [3] Department for Digital, Culture, Media and Sport (2018) [4] National Housing Federation (2018) [5] AI Growth Zones (2018) [6] Royal Town Planning Institute (2019) [7] The Alan Turing Institute (2018)

Artificial Intelligence's Role in Planning and Construction: A Friend or a Threat?