TY - JOUR
T1 - Creating visualizations using generative AI to guide decision-making in street designs: A viewpoint
AU - Valença, Gabriel
AU - Azevedo, Carlos M. Lima
AU - Moura, Filipe
AU - de Sá, Ana Morais
PY - 2025
Y1 - 2025
N2 - Architecture software tools are usually used to illustrate new street design layouts (e.g., Computer-Aided Design). However, these tools are not appropriate for the co-creation of street design solutions mainly due to the demanding work to create complex designs, the lack of multi-user interfaces, and the inability to create visualizations in real-time. Recently, a few generative AI tools such as UrbanistAI, PlacemakingAI, and Laneform have been developed to overcome these limitations, generating real-time street layout visualizations. These tools aim to enhance stakeholder and citizen involvement in street design processes by allowing citizens to easily modify street layouts and visualize how the street could be in the future. Even though these tools may increase efficiency in design generation, their possible impacts and integration into urban planning practices are poorly questioned and studied. This viewpoint aims to outline a research agenda, discussing the challenges and potential positive and negative effects of using generative AI in participatory decision-making for street designs. To the best of our knowledge, this is the first paper that discusses the possible benefits and impacts of these generative AI tools for generating future street design. We believe that integrating generative AI street design participation tools into urban planning processes has yet to be thoroughly understood, particularly in their impact on people's creativity and problem-solving, adaptability to different contexts, alignment with recent AI regulations, and implications for equity.
AB - Architecture software tools are usually used to illustrate new street design layouts (e.g., Computer-Aided Design). However, these tools are not appropriate for the co-creation of street design solutions mainly due to the demanding work to create complex designs, the lack of multi-user interfaces, and the inability to create visualizations in real-time. Recently, a few generative AI tools such as UrbanistAI, PlacemakingAI, and Laneform have been developed to overcome these limitations, generating real-time street layout visualizations. These tools aim to enhance stakeholder and citizen involvement in street design processes by allowing citizens to easily modify street layouts and visualize how the street could be in the future. Even though these tools may increase efficiency in design generation, their possible impacts and integration into urban planning practices are poorly questioned and studied. This viewpoint aims to outline a research agenda, discussing the challenges and potential positive and negative effects of using generative AI in participatory decision-making for street designs. To the best of our knowledge, this is the first paper that discusses the possible benefits and impacts of these generative AI tools for generating future street design. We believe that integrating generative AI street design participation tools into urban planning processes has yet to be thoroughly understood, particularly in their impact on people's creativity and problem-solving, adaptability to different contexts, alignment with recent AI regulations, and implications for equity.
U2 - 10.1016/j.urbmob.2025.100104
DO - 10.1016/j.urbmob.2025.100104
M3 - Journal article
SN - 2667-0917
VL - 7
JO - Journal of Urban Mobility
JF - Journal of Urban Mobility
M1 - 100104
ER -