Making Sense of Emerging Practice: AI in Planning Use Cases

Planners across the country are using artificial intelligence (AI) from large language models (LLMs) for daily tasks to the integration of AI into public-facing planning projects and outcomes. Use cases are proliferating rapidly, and it's hard to keep track of what planners are doing on the ground.

Making Sense of Existing and Emerging Practices

To make sense of existing and emerging practice, APA created the AI in Planning database that can be viewed as a map or a list, and houses more than 70 examples (and growing) of AI use in planning and local government practice both in the U.S. and abroad. It is structured according to planning practice areas, planning processes, type of AI technology involved (such as generative AI, machine learning, computer vision and natural language processing), and the role of AI in the workflow.

This tool helps planners track and understand the evolving landscape of AI use cases taking shape across different communities. Updated quarterly, the database is designed to capture the rapid pace and expansion of this technology in planning at the local and state level, and to do so without making value judgments, but simply to reflect reality.

AI in Action

The majority of collected use cases are in permitting and development and transportation, but the variety of cases shows that planners and other public agencies are using AI in all scopes of their work. AI-powered permitting platforms, which can speed up the approval process, are especially relevant for communities recovering from natural disasters. In transportation, AI-powered cameras are tracking illegal parking in bus lanes, creating digital traffic twins, and assessing road and sidewalk quality.

Sentiment Analysis

Public engagement is one of the largest, most time-consuming endeavors for planners. Planners must balance getting as much public input on projects and decisions as possible with associated time and monetary costs.

Bowling Green, Kentucky, understood this when undertaking the BG 2050 Comprehensive Plan. They wanted to ensure they received the most public input possible before making big decisions about the future of the city, which is set to double in size in the next 25 years. Subsequently, the city launched What Could BG Be?, a month-long online public polling and commenting forum supported by the Computational Democracy Project and Google Jigsaw that uses AI machine learning to pose questions to participants and to understand and synthesize massive amounts of community feedback into something plan makers can use effectively in the long-range strategic plan. All the data collected from the discussion can be viewed through the BG 2050 website, as can the team's content moderation strategies

In partnership with the University of Hawaii, the state's department of transportation is asking residents to install a free external camera on their vehicles as part of their Eyes on the Road initiative. AI is used to analyze the camera footage and will determine which roads are the most dangerous and require the most maintenance. The state is distributing 1,300 cameras to citizen volunteers across Oahu, Maui, Kauai, and Hawaii islands.

ADA Accessibility

Irvine, California, is employing AI-powered robots to map ADA accessibility of sidewalks across the city. The robots travel along set paths looking for criteria the city has identified that would determine a particular sidewalk to be inaccessible, such as tripping hazards and changes in sidewalk grade. These robots can collect data faster than city employees, and humans are able to spend more time on data analysis and decision-making.

There are significant limitations to note. Publicly available information represents only a portion of the AI-related work and experimentation occurring within local and state governments. Many pilots, internal tools, and early-stage initiatives are not formally documented or publicly shared. This limitation may be addressed through complementary methods such as in-depth interviews with practitioners or structured surveys to capture non-public insights.

Furthermore, AI deployment in the public sector is evolving quickly, meaning the database represents a snapshot in time. New tools, pilots, and policies may emerge shortly after data collection, requiring continuous updates to remain relevant. Further discussion on research limitations can be found in the methodology tab of the dashboard.

Next Steps

The AI landscape is changing at a rapid pace. This research is part of understanding how AI affects the planning profession. With time, it will be possible to see which applications are lasting and scalable, and which fall to the wayside. APA does not endorse or oppose planners using AI; instead, we acknowledge that the technology will be used in the field no matter what. Documenting these cases for members may help planners start conversations and learn from their peers. As local planning agencies and the municipalities that house them develop AI ethics and governance policies, understanding how AI might be applied in practice will help planners make more informed decisions.

AI is one of APA's key research priorities and part of a broader initiative to help planners navigate the changing environment triggered by AI's proliferation. Upcoming issues of Planning magazine and the Journal of the American Planning Association (JAPA) will explore responsible AI adoption in planning and raise other important questions for planners in this context.

We also value your feedback. If you know of a use case that belongs in this database, please send an email to Zhenia Dulko atzdulko@planning.org.

Top image: Illustration by Chiara Vercesi.