Cities worldwide are grappling with the mounting pressures of sustainability amid accelerating climate change, urban population growth, and shifting social dynamics. The design and regulation of urban environments directly influence how communities interact with their surroundings, how they move, and how resilient they become in the face of environmental stressors. Traditional zoning ordinances, particularly those prevalent across the United States, have historically prioritized segregation of land uses—separating residential, commercial, and industrial functions into discrete zones. This long-standing approach, while administratively straightforward, has inadvertently entrenched challenges related to vehicular dependence, sprawling development patterns, and fragmented social networks, undermining the holistic goals of contemporary sustainable urbanism.
In an ambitious new study, researchers Andrea Salazar-Miranda and Emily Talen harness artificial intelligence, specifically natural language processing (NLP), to analyze zoning codes across more than 2,000 census-designated places throughout the United States. Their work uncovers how municipalities are increasingly integrating form-based codes (FBCs) into their urban planning frameworks as a promising alternative to legacy zoning models. Unlike traditional zoning, FBCs emphasize the physical form and design of the built environment rather than segregating land uses, promoting compact, walkable, and mixed-use neighborhoods. These codes seek to foster environments where streets, buildings, and public spaces collectively support sustainable and equitable urban living.
The use of NLP in this context is a breakthrough, enabling the systematic examination of complex legal documents at scale. Zoning codes, often voluminous and written in inconsistent legal language, have historically been difficult to analyze comprehensively. By applying algorithms capable of parsing linguistic patterns and extracting relevant features indicative of FBC principles, the study achieves an unprecedented breadth of insight into how these codes are structured and implemented nationally. This methodological innovation not only enhances urban planning research but also paves the way for data-driven policy assessment and reform.
Findings from this large-scale analysis reveal that form-based codes have been widely adopted across diverse geographic regions in the United States but exhibit marked regional variations. Some metropolitan areas and smaller municipalities demonstrate more robust integration of FBCs, reflecting localized policy priorities and urban development histories. Importantly, the presence of form-based zoning correlates with key urban form indicators consistent with sustainable development: increased floor area ratios suggest greater density, while narrower, consistent street setbacks and smaller plot sizes indicate finer-grain urban fabric conducive to pedestrian activity and street-level engagement.
Beyond physical form, the study establishes strong associations between form-based codes and improved urban mobility outcomes. Places with FBCs report enhanced walkability scores, underscoring how built environment interventions translate into more accessible, human-scale neighborhoods. This shift away from automobile-centric design fosters shorter average commute times, reducing carbon emissions and supporting climate mitigation goals. Moreover, the increased share of multifamily housing in these areas signals a diversification of housing stock that can improve affordability and social inclusiveness.
The researchers articulate that the adoption of form-based codes represents a paradigm shift in zoning philosophy—from control of use to control of form. This subtle yet profound change addresses many shortcomings of conventional zoning that have contributed to urban sprawl, car dependency, and socio-spatial segregation. By focusing on the physical qualities of the urban environment such as building placement, street connectivity, and pedestrian realm quality, FBCs promote integrated communities where daily needs are accessible within walking distance, supporting healthier and more resilient urban lifestyles.
Salazar-Miranda and Talen’s work also highlights the critical potential for AI-enhanced tools to democratize access to zoning data and planning insights. Planners, policymakers, and community advocates can benefit from such analytical capabilities to benchmark existing codes, identify best practices, and cultivate evidence-based reforms. The approach underscores a future wherein machine learning and natural language processing augment human expertise, accelerating the transition toward sustainable urban governance.
One noteworthy insight from the study is that while form-based codes support equitable urban development, their implementation is not uniform. Regional disparities suggest underlying political, economic, and cultural factors that influence local zoning reform trajectories, pointing to the need for tailored strategies that respect contextual realities. For instance, cities in the Sun Belt may face different pressures—such as rapid suburban growth or climate vulnerabilities—compared to older, denser cities in the Northeast, impacting how FBC principles are applied and enforced.
The research also delves into the implications of form-based coding on housing markets, revealing that diversified urban forms encouraged by FBCs facilitate multifamily housing development. This has significant consequences for housing affordability and social diversity in cities frequently strained by rising costs and uneven access. By allowing compatible mixed uses and higher densities, form-based zoning can enable more inclusive neighborhoods that accommodate a range of socioeconomic groups, countering exclusionary patterns induced by traditional zoning barriers.
In terms of technical contributions, the study’s application of natural language processing involved a sophisticated pipeline to preprocess zoning texts, identify semantic clusters, and classify code components indicative of FBC guidelines. This NLP-driven framework extracted metrics such as references to building form, street frontage standards, and public space requirements, translating qualitative code language into quantifiable indicators. These metrics were then cross-referenced with spatial and demographic data to validate correlations with urban form and social outcomes, showcasing a rigorous multidisciplinary methodology.
The implications of this work extend beyond US cities. As urban areas globally confront sustainability imperatives, the lessons from adopting and assessing form-based codes through AI tools offer a replicable model. Cities elsewhere can explore analogous coding reforms to promote compactness, walkability, and mixed-use environments, addressing urban ailments from traffic congestion to environmental degradation. The portability of the NLP analysis approach can facilitate comparative studies and knowledge transfer across diverse geographic contexts.
The increased compactness and walkability associated with form-based codes also have direct public health benefits. By encouraging active transportation modes such as walking and cycling, these zoning reforms contribute to reduced air pollution, enhanced physical activity, and improved mental well-being. This multidimensional impact underscores how urban design, regulatory frameworks, and technological innovation collectively influence the quality of life at the community scale.
In conclusion, the integration of artificial intelligence in urban planning research, as demonstrated by Salazar-Miranda and Talen, sheds illuminating light on zoning reforms that hold promise for transforming cities into more sustainable, livable places. The widespread yet uneven adoption of form-based codes in the United States exemplifies the ongoing evolution of regulatory tools to better align with environmental imperatives and social equity goals. By quantifying the influence of these codes on urban form, mobility, and housing, the study builds a compelling case for broader adoption and continued refinement of form-based approaches.
This research not only advances theoretical understanding but also offers practical guidance for municipalities seeking to modernize their zoning frameworks. Planners and policymakers are encouraged to leverage AI-enabled analytics to diagnose existing code deficiencies and strategize reforms that nurture compact, integrated, and resilient urban ecosystems. The fusion of technological innovation and planning expertise represents a pivotal path forward as cities navigate the challenges of the 21st century and strive toward a sustainable urban future.
Subject of Research: Zoning reforms and their impact on urban sustainability, with a focus on form-based codes analyzed via natural language processing across US municipalities.
Article Title: An AI-based analysis of zoning reforms in US cities.
Article References:
Salazar-Miranda, A., Talen, E. An AI-based analysis of zoning reforms in US cities. Nat Cities 2, 304–315 (2025). https://doi.org/10.1038/s44284-025-00214-0
Image Credits: AI Generated
