In that Empire, the Art of Cartography attained such Perfection that the map of a single Province occupied the entirety of a City, and the map of the Empire, the entirety of a Province. In time, those Unconscionable Maps no longer satisfied, and the Cartographers Guilds struck a Map of the Empire whose size was that of the Empire, and which coincided point for point with it. The following Generations, who were not so fond of the Study of Cartography as their Forebears had been, saw that that vast Map was Useless, and not without some Pitilessness was it, that they delivered it up to the Inclemencies of Sun and Winters. In the Deserts of the West, still today, there are Tattered Ruins of that Map, inhabited by Animals and Beggars; in all the Land there is no other Relic of the Disciplines of Geography.
Jorge Luis Borges, On Exactitude in Science, 1946
The desire for a complete representation of reality is nothing new. Long a fantasy of cartographers, scientists, and business titans, the promise of a containable, fully knowable reality has inspired confident exploration, with its corresponding military, market, or scientific conquest. As Borges makes clear in his 1946 parable, ‘On the Exactitude in Science’, the hubris of one generation is the regret of the next.1 Jorge Luis Borges, “On Exactitude in Science,” in Collected Fictions, trans. Andrew Hurley (Penguin, 1998), 325. Imagine a map of a city that was the exact size of that city, with every spatial detail fully contained within its representation. Almost immediately, the distinction between the place and the map would fall away. The city would be rendered fixed and legible, the complexities of its interlocking systems fully knowable to those within it. So why, in Borges’ tale, did future generations grow less fond of the study of cartography? Because there would no longer be a role for future generations. If the map locked space to time, nothing would change. The temporal dynamics of the city would be frozen, and to experience the city after the map was created would merely be a re-presentation of a past ‘reality’.
In 2026, the race to develop the complete map has only accelerated. In the context of rapid datafication and the capacity of algorithmic comprehension, the smart city movement has led to a geopolitical competition2Martin Bugaj, Veronika Machova and Sofia Bratu, “The Geopolitics of Smart Sustainable City Governance and Simulated 3D Environments: Urban Digital Twins, Sensor Data Processing Algorithms, and Extended Reality Technologies,” Geopolitics, History, and International Relations 14, no. 2 (2022): 152-167, https://www.jstor.org/stable/48710618.—something resembling a space race— to make the city fully knowable. These efforts, from Toronto to Seoul to Delhi, have focused on the interoperability of systems, the proliferation and legibility of data, and the enhanced efficiency of services. While the dominant vision of the contemporary smart city is more dynamic than Borges’ discarded map, it too often shares the hubristic desire for completeness that is more likely to restrict social and cultural mobility than expand it.3Donatus Ebere Okonta and Vladimir Vukovic, “Smart cities software applications for sustainability and resilience,” Heliyon 10, no. 12 (2024): https://doi.org/10.1016/j.heliyon.2024.e32654.
But it doesn’t have to be this way. In his 2002 book Emergence, Steven Johnson uses a different metaphor.4Steven Johnson, Emergence: The Connected Lives of Ants, Brains, Cities, and Software (Scribner, 2002).. He describes the city as a collective brain—its intelligence an emergent property stemming from interactions between residents and their environment. It is plastic, generative, constantly evolving from birth to death. The city is a collection of different intelligences—from spatial patterns, to social interactions, to cultural expressions. Cities (taken as a whole) are not mere representations, they are complex systems capable of learning from the past, present, and future. Generative Artificial Intelligence (AI), a technology that is itself ubiquitous, has within it the possibility to contain, but it can also expand to include more flexible urban knowledge, a way of seeing cities that is both coherent and emergent. In this spirit, we have launched the Urban Intelligences Project (UIP) at Boston University (BU) to explore next-generation urban cartography that brings together complex humanistic knowledge with advanced computational capabilities.
In this short essay, we describe the conceptualization and design of an AI-enabled urban digital twin (UDT) of one of Boston’s inner-city neighborhoods. Our focus is on the Fenway neighborhood—famous as home to the Boston Red Sox’s Fenway Park and where BU is located. Positioned just west of downtown, Fenway is populated by young professionals, medics, and college students (the Meds and Eds Boston is known for); it contains the US’s oldest WWII Victory Garden and Back Bay Fens; and it is one of Boston’s fastest-evolving districts, with luxury high-rises, lab spaces, and Time Out Market. Despite rapid development and gentrification, the Fenway Community Development Corporation, as well as many other community-based organizations, supports a diverse population through health care services and housing affordability efforts.
A UDT is an integrated, spatio-temporal model of a city’s physical and socio-technical systems often used for simulation, prediction, or decision-making. They tend to fall into one of three categories: spatial (a 3D model of an urban space); informational (a convergence of building/city information modeling, and/or geographical information systems); or more recently, transitional (a dynamic representation augmented with real-time data on traffic, air quality, or other urban flows).5Imane Jeddoub, Ghislain-Amaury Nys, Rafika Hajji and Roland Billen, “Digital Twins for Cities: Analyzing the Gap between Concepts and Current Implementations with a Specific Focus on Data Integration,” International Journal of Applied Earth Observation and Geoinformation 122 (2023): 103440, https://doi.org/10.1016/j.jag.2023.103440. Following Kitchin and Dawkins, the Fenway Digital Twin (FDT) combines all three. It is “not static, but dynamic – viewable from different perspectives and changing in sync with alterations in the geographic space.”6Rob Kitchin and Oliver Dawkins, “Digital Twins and Deep Maps,” Transactions of the Institute of British Geographers 50 (2024): e12699, https://doi.org/10.1111/tran.12699. As such, it moves beyond standard spatial flows, providing a more dynamic understanding of place.
Digital Twins are broadly understood as realistic digital representations of physical ‘things.’7Michael Batty, “Digital Twins,” Environment and Planning B: Urban Analytics and City Science 45, no. 5 (2018): 817–820, https://doi.org/10.1177/2399808318796416. Given their ability to gauge the pulse of a city, they can serve well as tools for urban decision making. Cities are among the most complex systems humanity has produced—layered accumulations of social memory, built form, political struggles, ecological processes, and everyday life. Yet despite rapid advances in artificial intelligence and the proliferation of urban data, computational models of cities remain impoverished in humanistic depth. Current AI-driven urban systems—smart cities, planning tools, digital twins—optimize for efficiency, throughput, or quantifiable outcomes and systematically exclude the cultural, historical, social, and ecological dimensions that animate what neighborhoods mean to the people who live in them,8Timothy Moss, Friederike Voigt and Sören Becker, “Digital urban nature: probing a void in the smart city discourse,” City 25, no. 3-4, (2021): 255-276, https://doi.org/10.1080/13604813.2021.1935513; Yanhua Yao , Zheyu Li and Sai Ma, “Exploring Gamification’s Role in Shaping Socially Sustainable Urban Spaces: A Case Study of Gensen in SOLANA, Beijing,” Buildings 15, no. 12 (2025): 1969, https://doi.org/10.3390/buildings15121969. Not merely a technical gap, this omission reflects a deeper epistemological divide between the sciences of the city and the humanities of the city—a divide with real consequences for determining whose history is preserved and whose future is planned.
The FDT will catalyze a new form of dynamic humanistic inquiry that activates knowledge through interaction and makes knowable the relational fabric of urban life. The challenge is how to represent human urban intelligences using multiple AI agents to synthesize the distributed, contested, multi-temporal knowledges of a living neighborhood into an interactive, evolving representation. We address a dual gap: the humanities lack scalable computational tools to synthesize and activate the complex, multi-sited knowledge they produce, and Large Language Models (LLMs) tend to lack the cultural grammar needed to represent urban environments with the nuance and accountability that communities deserve.
We converge two disciplinary questions. From the humanities: How can the distributed, contested, and layered intelligences of an urban neighborhood be captured, formalized, and activated in a dynamic AI representation that is accountable to the community it represents? This is fundamentally a question about urban epistemology: what does it mean to ‘know’ and understand a city, and who gets to define that knowledge? AI offers an unprecedented instrument to synthesize the vast, heterogeneous data—archival, ethnographic, environmental, architectural—that humans produce about cities, and to make that knowledge interactive, accessible, and generative in ways that print scholarship cannot. From AI: How can multi-agent AI systems be designed to encode and dynamically represent culturally specific, humanistically grounded knowledge drawn from community engagement, and how can agents meaningfully interact with one another to model the relational complexity of an urban neighborhood? The AI models underlying current multi-agent systems are typically trained on structured, quantitative, or generic text data and are not designed to hold or reason about cultural, historical, or environmental particularity. The FDT will use complex cultural data (quantitative and qualitative data related to arts, humanities, heritage, and human social behavior, eg. digital archives, museum collections, 3D scans of archaeological sites, minutes of community meetings, etc.) to develop a framework capable of representing places with specificity and depth. Humanistic expertise is essential to this framework that moves beyond the default surface-level pattern matching that strips the cultural content that matters most.
Participatory methods to expand the right to the digital twin city
“Researchers need to explore bottom-up models and forms of citizen-led urban designs and models that can inform the quest to achieve the Right to the Digital Twin City.”9Arthur De Jaeger and Thomas Swerts, “Right to the Digital Twin City? Citizen Participation and Limits-by-Design in Rotterdam’s Urban Digital Twin,” Cities 168 (2026): 106498, https://doi.org/10.1016/j.cities.2025.106498.
The paradox of digital twins is that technological advancements have progressed with disconnection from human participation. Our interest is to create a digital twin that effectively combines the spatial, informational, and transitional qualities of UDTs by dynamically incorporating humanistic knowledge. It will take an iterative community-embedded approach to building a multi-agent digital representation of the Fenway neighborhood by incorporating communities in the loop (CITL) at every stage of data collection and model development, ensuring that technical decisions remain accountable to community values.10Yiwei Wang, Sai Raj Nagireddy, Chandra Teja Thota, Do Hyung Ho and Yoonchang Lee, “Community-in-the-Loop: Creating Artificial Process Intelligence for Co-production of City Service,” Proceedings of the ACM on Human-Computer Interaction 6, no. CSCW2 (2022): 1–21, https://doi.org/10.1145/3555176. To date most UDT development has not included active forms of participation—such as community or co-design.11Andrew Wilson and Marc Tewdwr-Jones, Digital Participatory Planning: Citizen Engagement, Democracy, and Design (Routledge, 2021).. This lack of focus on embedding participation in the design and implementation of urban analytics tools is addressed head-on in the FDT. Existing evidence on successful citizen engagement in developing urban digital twins is scarce, so we are developing our own strategy that addresses the unequal power dynamic between ourselves as shapers and producers of the digital twin and local citizens.12Jaume Ferré-Bigorra, Miquel Casals and Marta Gangolells, “The Adoption of Urban Digital Twins,” Cities 131 (2022): 103905, https://doi.org/10.1016/j.cities.2022.103905. Like De Jaeger and Swerts we believe in the ‘Right to the Digital Twin City’, the “collective right of inhabitants to exercise citizen power over processes of digital—and physical—urbanization by making and remaking themselves and their cities in the urban digital twin environment.”13De Jaeger and Swerts, “Right to the Digital Twin City?” Following their suggestion we involve residents “at the earliest stages of the design phase, extending beyond mere input to active participation in shaping the process itself.”14Wang et. al., “Community in the Loop.”
Our process includes identifying relevant “sites” in the Fenway neighborhood (those 25 buildings, spaces, and streets noted above that will be converted into agents) by convening community conversations (structured group dialogues designed to bring residents, leaders, and organizations together) that will look at and discuss sites that community members identify as significant—the local (human) meanings of urban space and place in Fenway. Our goal is to bring diverse local knowledges and experiences ‘situated’ in Fenway to the table, being as inclusive as possible of all those in the Fenway neighborhood. Over the course of several months, our research team will host approximately six community conversations in community spaces. Instead of university researchers calling the meetings, we will organize the conversations in partnership with established community-based organizations with long-term, trusted relationships in the community. The conversations will be approximately two hours long and focus on surfacing the places, issues, and experiences that matter most to residents. Participants will be compensated for their time in the conversations and invited to stay involved to review and validate the insights produced through the conversations. A list of the 25 priority sites will be circulated to all participants before being finalized. Sites might include Fenway Park, a baseball stadium with its layered history of cultural significance (see figure 1), or the Fens Community Garden (see figure 2), with its turbulent formation and the lifestyle benefits it provides to residents.
Identifying locations and their relevant social groups
The process of identifying which 25 locations or historical sites matter most to a community cannot be determined by experts alone—it requires what Costanza-Chock calls ‘a design justice approach’, in which the community itself is positioned not as a passive recipient of decisions but as the primary author of them.16Sasha Costanza-Chock, Design Justice: Community-Led Practices to Build the Worlds We Need (MIT Press, 2020). Drawing on the logic of participatory budgeting, where residents deliberate collectively over how shared resources are allocated, our process invites community members into structured yet open-ended conversations about collective memory, belonging, and place. Rather than producing a list handed down by historians or municipal planners, the process works through multiple rounds of community input, prioritization exercises, and public deliberation, ensuring that the final selections reflect a genuinely plural sense of what is worth preserving and why. Meaningful civic participation depends on designing engagements that are not merely consultative but constitutive—where participants experience their input as materially shaping outcomes.17Eric Gordon and Paul Mihailidis, eds., Civic Media: Technology, Design, Practice (MIT Press, 2016). Costanza-Chock’s insistence on centering those most impacted, particularly communities historically excluded from heritage and preservation decisions, further frames the process as one of repair as much as documentation: the goal is not simply to inventory the past but to redistribute the power to define it.
Once the locations are identified by the community, we will elevate what Bijker et al. refer to as relevant social groups.18Wiebe E. Bijker, Thomas Parke Hughes and Trevor Pinch, eds., The Social Construction of Technological Systems (MIT Press, 2012). These are groups of actors who coalesce around particular technological frames that reflect their shared interpretations of a technical artefact. We will engage these groups specifically to construct each of the AI agents. Of course, there will always be fallibility in that certain groups might be excluded, however unwittingly, from the participatory process, and ultimately from the layers of understanding. To mitigate this risk, we will practice diverse recruitment and outreach, we will partner with leaders of community-based organizations, and we will take the time to extend participation in community conversations both wide and deep.
Despite these efforts, there is still the possibility that the FDT will not be seen as inclusive or representative of the neighborhood. This is true for both procedural and structural reasons. Regarding procedure, there will always be people who feel like they were not consulted, or were consulted insufficiently, and as a result may approach the FDT with suspicion. Regarding structure, the FDT is inherently incomplete. It will be composed of only 25 sites in the neighborhood, and through a process of prioritization, will necessarily exclude sites. Sites might be excluded by oversight or intention. Oversight can result from gaps in the process, and unintentional exclusions of stakeholder voices. And intention can result from the community’s deliberate desire to exclude certain sites, because they are deemed unimportant or even politically dangerous. The research team will pay attention to the power and politics associated with both oversight and intention. However, guided by Borges’ warnings of the dangers of completeness, part of the exclusion is by design. The FDT is not just a compilation of nodes, but nodes in relation to each other, where the whole is more than the sum of its parts. The weaving together of urban intelligences requires sufficient breathing room, it requires the creation of dialogic space where ways of knowing are represented as incomplete and plural, and so therefore dependent on human interaction.
AI agents with domain knowledge
Once a location and its social group have been identified, we will construct a dedicated AI agent that functions as a living repository of that place’s layered history and present condition. The process begins with the systematic collection of publicly available structured data—911 and 311 call records, municipal permits and inspections, zoning decisions, policy documents, and planning reports—alongside newspaper archives that capture how the location has been covered, contested, and narrated over time. This documentary record is then enriched with oral histories gathered directly from current residents, whose accounts introduce forms of knowledge that official records routinely omit: the texture of daily life, patterns of change that preceded formal documentation, and the relational significance of spaces that data alone cannot convey. Together, these sources are assembled into a curated corpus specific to the location.
The AI agent is then built using an approach known as agentic retrieval-augmented generation (agentic RAG). Rather than retraining a general-purpose language model on the corpus, an agentic RAG system keeps the corpus in a searchable database and allows the model to retrieve from it dynamically at query time, generating its own search queries through iterative reasoning.19Patrick Lewis, Ethan Perez, Aleksandra Piktus et al., “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks,” Advances in Neural Information Processing Systems 33 (2020): 9459–9474, https://dl.acm.org/doi/abs/10.5555/3495724.3496517. This architecture is particularly well-suited to this project. Because each response is grounded in retrieved source documents, the agent can reason from the specific oral history transcript, permit record, or newspaper article that informed it. It also allows the corpus to evolve: as new oral histories are collected or community members request changes, the underlying database can be updated without retraining the model. Moreover, it substantially reduces the risk of hallucinated content as it is grounded on human-verified data. This method allows the model to prioritize knowledge particular to that location: its rhythms, its conflicts, its institutions, and its voices. The result is an agent that retains general reasoning capacity while becoming, in effect, a specialist in that place—capable of responding to queries, surfacing connections across sources, and reflecting the perspectives embedded in both the archival record and the lived experience of those who call it home.
Grounding the models to interact
While each agent is aligned to function as a deep specialist in a single location, the architecture’s most powerful analytic capacity emerges when the agents are allowed to communicate with one another—querying, comparing, and synthesizing across the full network of 25 sites simultaneously. In a multi-agent configuration, individual agents do not merely answer questions in isolation; they can be prompted to collaborate on compound queries that no single agent could resolve alone, producing what complexity theorists would describe as emergent knowledge: insights that arise from the interaction of the system’s parts rather than from any one of them. A question about displacement, for instance, might cause the agent grounded on a rapidly gentrifying commercial corridor to determine patterns in permitting and inspection data that, when compared against the responses of agents grounded on adjacent locations, reveal a coordinated logic of disinvestment and reinvestment that would be invisible from any single vantage point. Similarly, agents can be prompted to identify temporal resonances across sites, moments when similar conditions appeared in different neighborhoods at different times, effectively allowing the network to model the city as a dynamic and interconnected system rather than a collection of discrete places. This is consistent with Shannon Mattern’s argument that cities are best understood as layered, communicative infrastructures whose meaning is produced relationally, across scales and over time.20Shannon Mattern, “A City Is Not a Computer,” Places Journal (February 2017), accessed April 17, 2026” https://doi.org/10.22269/170207. By enabling the agents to speak to one another, the system moves beyond archive and toward analysis, offering urban planners, community organizations, and researchers a tool capable of surfacing the structural patterns of investment, neglect, contestation, and change that shape neighborhood life but rarely appear in any single dataset.
Conclusion
The FDT is far from the abstract, objective, rationalizing, and disembodied digital twin that Gillian Rose warns against in her recent article describing the deleterious effects of hyper-rational urban models.21Gillian Rose, “Visualising Human Life in Volumetric Cities: City Digital Twins and Other Disasters,” Dialogues in Urban Research 3, no. 2 (2024): 148-169, https://doi.org/10.1177/27541258241288812. Our digital twin recognizes people as sensors, enablers, and actuators, and acknowledges the vital, multi-faceted role of residents in shaping sustainable urban societies. The FDT is not yet built. It is a work of speculative fiction that, even if not fully realized as described, paints a picture of a computational urban future that can unlock humanity’s potential rather than containing it. Of course, our aspiration is to build the FDT, and for the City of Boston’s Planning Department to adopt it as part of its decision-making process. But even if that never happens, the conceptual model of the FDT should inspire city agencies to adopt democratic and inclusive procedures for incorporating human experience into computational urban decision-making.22Timea Nochta and Kwadwo Oti-Sarpong, “Participation matters: The social construction of digital twins for cities,” Environment and Planning B: Urban Analytics and City Science 53, no. 4 (2024): 765-777, https://doi.org/10.1177/23998083241305695. Heeding Borges’ warning about the harms of a total and static map of the city and region, in our current moment of ballooning technological enthusiasm and capability, we must imagine and build dynamic systems that contain and inspire a multiplicity of urban intelligences.
Notes
| ↑1 | Jorge Luis Borges, “On Exactitude in Science,” in Collected Fictions, trans. Andrew Hurley (Penguin, 1998), 325. |
|---|---|
| ↑2 | Martin Bugaj, Veronika Machova and Sofia Bratu, “The Geopolitics of Smart Sustainable City Governance and Simulated 3D Environments: Urban Digital Twins, Sensor Data Processing Algorithms, and Extended Reality Technologies,” Geopolitics, History, and International Relations 14, no. 2 (2022): 152-167, https://www.jstor.org/stable/48710618. |
| ↑3 | Donatus Ebere Okonta and Vladimir Vukovic, “Smart cities software applications for sustainability and resilience,” Heliyon 10, no. 12 (2024): https://doi.org/10.1016/j.heliyon.2024.e32654. |
| ↑4 | Steven Johnson, Emergence: The Connected Lives of Ants, Brains, Cities, and Software (Scribner, 2002). |
| ↑5 | Imane Jeddoub, Ghislain-Amaury Nys, Rafika Hajji and Roland Billen, “Digital Twins for Cities: Analyzing the Gap between Concepts and Current Implementations with a Specific Focus on Data Integration,” International Journal of Applied Earth Observation and Geoinformation 122 (2023): 103440, https://doi.org/10.1016/j.jag.2023.103440. |
| ↑6 | Rob Kitchin and Oliver Dawkins, “Digital Twins and Deep Maps,” Transactions of the Institute of British Geographers 50 (2024): e12699, https://doi.org/10.1111/tran.12699. |
| ↑7 | Michael Batty, “Digital Twins,” Environment and Planning B: Urban Analytics and City Science 45, no. 5 (2018): 817–820, https://doi.org/10.1177/2399808318796416. |
| ↑8 | Timothy Moss, Friederike Voigt and Sören Becker, “Digital urban nature: probing a void in the smart city discourse,” City 25, no. 3-4, (2021): 255-276, https://doi.org/10.1080/13604813.2021.1935513; Yanhua Yao , Zheyu Li and Sai Ma, “Exploring Gamification’s Role in Shaping Socially Sustainable Urban Spaces: A Case Study of Gensen in SOLANA, Beijing,” Buildings 15, no. 12 (2025): 1969, https://doi.org/10.3390/buildings15121969. |
| ↑9 | Arthur De Jaeger and Thomas Swerts, “Right to the Digital Twin City? Citizen Participation and Limits-by-Design in Rotterdam’s Urban Digital Twin,” Cities 168 (2026): 106498, https://doi.org/10.1016/j.cities.2025.106498. |
| ↑10 | Yiwei Wang, Sai Raj Nagireddy, Chandra Teja Thota, Do Hyung Ho and Yoonchang Lee, “Community-in-the-Loop: Creating Artificial Process Intelligence for Co-production of City Service,” Proceedings of the ACM on Human-Computer Interaction 6, no. CSCW2 (2022): 1–21, https://doi.org/10.1145/3555176. |
| ↑11 | Andrew Wilson and Marc Tewdwr-Jones, Digital Participatory Planning: Citizen Engagement, Democracy, and Design (Routledge, 2021). |
| ↑12 | Jaume Ferré-Bigorra, Miquel Casals and Marta Gangolells, “The Adoption of Urban Digital Twins,” Cities 131 (2022): 103905, https://doi.org/10.1016/j.cities.2022.103905. |
| ↑13 | De Jaeger and Swerts, “Right to the Digital Twin City?” |
| ↑14 | Wang et. al., “Community in the Loop.” |
| ↑15 | Eric Gordon, Steven Schirra and Justin Hollander, “Immersive Planning: A Conceptual Model for Designing Public Participation with New Technologies,” Environment and Planning B: Planning and Design 38, no. 3 (2011): 505–519, https://doi.org/10.1068/b37013; Eric Gordon, John Harlow, Melissa Teng and Elizabath Christoferetti, “Toward a Collaborative Smart City: A Play-Based Urban Living Laboratory in Boston,” International Journal of Human–Computer Interaction 39, no. 2 (2023): 302–318, https://doi.org/10.1080/10447318.2021.2012384. |
| ↑16 | Sasha Costanza-Chock, Design Justice: Community-Led Practices to Build the Worlds We Need (MIT Press, 2020). |
| ↑17 | Eric Gordon and Paul Mihailidis, eds., Civic Media: Technology, Design, Practice (MIT Press, 2016). |
| ↑18 | Wiebe E. Bijker, Thomas Parke Hughes and Trevor Pinch, eds., The Social Construction of Technological Systems (MIT Press, 2012). |
| ↑19 | Patrick Lewis, Ethan Perez, Aleksandra Piktus et al., “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks,” Advances in Neural Information Processing Systems 33 (2020): 9459–9474, https://dl.acm.org/doi/abs/10.5555/3495724.3496517. |
| ↑20 | Shannon Mattern, “A City Is Not a Computer,” Places Journal (February 2017), accessed April 17, 2026” https://doi.org/10.22269/170207. |
| ↑21 | Gillian Rose, “Visualising Human Life in Volumetric Cities: City Digital Twins and Other Disasters,” Dialogues in Urban Research 3, no. 2 (2024): 148-169, https://doi.org/10.1177/27541258241288812. |
| ↑22 | Timea Nochta and Kwadwo Oti-Sarpong, “Participation matters: The social construction of digital twins for cities,” Environment and Planning B: Urban Analytics and City Science 53, no. 4 (2024): 765-777, https://doi.org/10.1177/23998083241305695. |
Eric Gordon is faculty in the College of Communication and the director of the Center for Media Innovation & Social Impact at Boston University. He studies civic technology and public participation in cities, and is the author (or editor) of five books and over 50 articles on participatory urban technology, including The Urban Spectator (Dartmouth, 2010), Net Locality (Blackwell, 2011), Civic Media (MIT Press, 2016), Ludics (Palgrave, 2021), and Meaningful Inefficiencies (Oxford, 2020). In addition to his scholarship, he has led dozens of collaborative design projects all over the world that bring the public and civic sectors together to foster democratic process with technology. His forthcoming book How Institutions Listen: Civic Data, AI, and the Path To Public Trust will be published by MIT Press in 2026.
Professor Loretta Lees (FAcSS, MAE) is Director of the Initiative on Cities at Boston University. She is an urban geographer, urbanist, and urban social theorist who researches gentrification, urban regeneration, global urbanism, urban policy, and urban public space. Since 2024 she has co-organized the Boston Urban Salon; she previously co-organized 2009-2022 The Urban Salon: A London forum for architecture, cities and international urbanism and 2016-22 the Leicester Urban Observatory. Her awards include the 2026 Patron’s Medal, Royal Geographical Society and the 2025 Contribution to the Field of Urban Affairs Award, Urban Affairs Association.
Carolyn L. White is director of the Preservation Studies Program and Professor of History of Art and Architecture at Boston University. Her scholarship focuses on cultural heritage, the materiality of daily life, the built environment, active site archaeology, and the intersection of and collaboration between art and archaeology. Her most recent books are The Archaeology of Burning Man (University of New Mexico Press, 2022) and Distant Voices: On Steven Seidenberg’s Architecture of Silence (Contrasto Press, 2024), the latter with her frequent collaborator, photographer Steven Seidenberg. Her current research focuses on squatted buildings and the right to the city in Rome, Italy.
Chang Xiao is an Assistant Professor in the Department of Computer Science at Boston University. His research explores how people can effectively steer and collaborate with increasingly capable AI systems, drawing on human-computer interaction, artificial intelligence, visual computing, and AR/VR. He develops human-centered interaction techniques and interfaces that connect human intention, computational capabilities, and the physical world. Previously, he was a Research Scientist at Adobe Research and earned his Ph.D. in Computer Science from Columbia University.
