Could AI advancements help the government hit housing targets?
As technology continues to evolve, Pete Canavan of Carter Jonas explains that it could be the key to achieveing the governments goal to building 1.5 million homes over the course of the next Parliament.
Despite ambitious housing targets, delays to existing plans remain. Research shows that seven English local planning authorities (LPAs) with draft plans at an advanced stage of preparation may be told to ‘go back to the drawing board’ because the gap between their proposed home targets and Labour’s revised housing need assessment is too great. At the same time, others are accelerating their plans to attempt to utilise ‘transitional arrangements’ and avoid housing need increases in the short term.
The planning system is, unfortunately, part of the problem. The current system works well in theory: nationally led policy on the broader issues (the NPPF), strategic planning on a local level to determine the allocation of development (local plans), local input on the siting of a development (neighbourhood planning), the masterplanning and detailed consideration of new communities (planning applications) and, as necessary, the appeal system, all provide a good structure for delivery.
However, issues arise when the system fails, mainly due to staffing shortages. The pressure on local authorities to prepare comprehensive local plans is immense. Local authorities have responsibility for everything from taxes to bin collection and are universally under-resourced.
Engagement with those who live and work in the area for which a strategic plan is being created is vital. This engagement must be regular, iterative, and relevant. The whole process of strategic planning must be a shared and transparent exercise, including the ‘why,’ the ‘where’ and the ‘how.’ Critically, the interrelationships between each of those three questions must also be explored, and also the relationships between the potentially competing priorities in the ‘planning balance’. A strategic planning engagement exercise must focus on delivering an output, rather than identifying how many people and organisations favour one ‘topic’ or ‘theme’ over another.
Increasing efficiency
So how can the planning process be made more efficient? In theory AI could greatly improve efficiency and decrease cost: generating and analysing housing or employment projections, reviewing and categorising site submissions, managing consultation and even auto-generating reports and analyses. But what more can be done?
The DLHUC’s PropTech engagement fund is already being used by 13 local authorities across the country to pilot the use of AI to manage public consultation on Local Plans. Authorities have adopted technology in several ways: for instance, Greater Cambridge analysed social media feedback that wasn’t captured on the consultation portal, and Southampton used 3D models to show how new proposals would look. AI can quickly review consultation responses and automatically categorise them, picking out key themes and identifying trends.
At the other end of the scale, could AI help to reviewing minor planning applications? Householder applications, Certificates of Lawfulness or conditions discharge are for the most part relatively simple but take up a great deal of officer time. This could in theory be automated by a computer program, with a planning professional only required to review the final recommendation. Similarly, simple pre-
application enquiries for small scale development could be automated with a chat bot, so applicants would interact with their planning department in the same way they would with their bank or mobile phone provider.
Proceed with caution
However, there may be potential downsides. For example, although automation could aid public engagement by targeting more specific groups on a Local Plan or application consultation, a computer’s perception of interests could lead to an artificial narrowing of options, or reinforcement of filter bubbles. In addition, AI has no intrinsic agency as it must be told what to do. Furthermore, it has no accountability, as its output must be evaluated by an accountable human. Care must be taken to ensure that consultation responses have been summarised correctly and the auto-generated parts of a report make sense. Where have the data used in models or reports come from? Are there inaccuracies? Is it replicating unintended biases?
I do not believe planning in the UK should ever be a ‘tick-box’ exercise. Planning relies on the exercise of judgement and weighing up the planning balance. Considerations of design or the impact of a proposal on heritage assets is subjective. Applicants and officers need room for discussion on where trade-offs or improvements can be made, and where departures from planning policies can be justified. And of course, decisions must have some kind of democratic oversight to ensure public good is balanced against private interest.
There is no doubt that AI has enormous potential to transform the way that data-driven and administrative tasks are undertaken, which could ease workloads and allow more time for planning (as opposed to administration). However, negotiating good planning outcomes will always require human intervention to exercise nuance, common sense, creativity and critical judgement.
In related news:
Letting agents could lose almost £400m under Renters’ Rights Bill