How AI Shapes Urban Planning: A Historical View

Author(s): Scott Douglas Jacobsen

Publication (Outlet/Website): The Good Men Project

Publication Date (yyyy/mm/dd): 2025/07/02

Public policy researcher Dr. Daniel Wortel-London explores how AI reshapes urban planning by echoing historical growth patterns, inequality, and innovation. Drawing on his forthcoming book The Menace of Prosperity, he compares the rise of AI to earlier technological shifts like highway expansion and the postwar white-collar economy. While AI offers tools for participatory budgeting and data-driven decisions, it also risks reinforcing existing biases. Dr. Wortel-London urges communities to engage early in AI governance and to prioritize inclusive, equitable growth—challenging the myth that all growth is inherently good or sustainable.

Scott Douglas Jacobsen: Today, we are speaking with Dr. Daniel Wortel-London. He is a public policy researcher and historian currently serving as a Visiting Assistant Professor at Bard College. His work focuses on twentieth-century economic and urban policy and the historical roots of inequality and economic development in New York City. He is completing a book titled The Menace of Prosperity: The Rise and Fall of the American City, 1865–1981, forthcoming from the University of Chicago Press later this year.

His scholarship has appeared in top academic journals and media outlets, including The Washington Post and Jacobin. A recipient of prestigious fellowships, Dr. Wortel-London has taught at institutions such as NYU and CUNY and held roles in policy research and museum curation. He is based in New York City. Thank you for joining me today. Let me find my notes—here they are. So, how is AI being integrated into urban planning initiatives?

Dr. Daniel Wortel-London: From the nineteenth century on, cities have used new technologies—from census-taking to social statistics—to promote growth and livability. AI continues this trend, though its adaptive, learning capabilities make it especially powerful—and potentially disruptive

A century ago, efforts were made to revolutionize municipal data management. Innovations in census-taking and the development of social statistics were used to plan highways, public housing, and other infrastructure. We’re seeing a scaled-up version that applies to new domains like urban sustainability. Of course, contemporary concerns—such as data privacy—also have historical parallels.

From a historical standpoint, today’s integration of AI reflects a long-standing goal of cities: to achieve economic growth, improve transportation and administrative efficiency, and enhance overall livability. My forthcoming book focuses on this evolution from 1865 to 1981. AI, of course, is a different kind of technology—it’s not just a tool like a calculator or database. It’s adaptive, autonomous in specific contexts, and capable of learning and evolving, making it powerful and potentially disruptive.

Jacobsen: How is artificial intelligence applied in urban contexts beyond planning? You mentioned historical parallels—are there similarities in the types of societal adaptations required now compared to those between 1865 and 1981?

Wortel-London: Absolutely. Cities have historically had to adapt to shifting environmental and infrastructural demands. For example, New York needed to secure fresh water in the mid-nineteenth century as its population rapidly expanded. Later, new management systems and public investments addressed challenges like traffic congestion, overcrowded housing, and public health crises. These responses often operated within constraints—many of which persist today. Chief among them is the emphasis on economic growth.

For instance, proposals to solve environmental problems are frequently limited by concerns over property values or the risk of wealthier residents and businesses relocating. That was true in the nineteenth century during the construction of the Croton Aqueduct system and remains true now in debates around renewable energy infrastructure and circular economy models.

In short, adaptability has always been essential to urban life. Cities are dynamic organisms, constantly changing in response to internal and external pressures. My book tries to explain how certain assumptions are baked into the DNA of many cities—assumptions that often persist and constrain our responses to challenges. These assumptions can prevent us from thinking more creatively about how to build vibrant, livable cities.

Of course, possibilities always have to adapt to economic conditions. One specific example from the early twentieth century is the introduction of the car—the Model T—and the associated transformations. That shift replaced the horse as the primary mode of transportation. There was even a whole industry built around cleaning up horse manure, which had been a major urban problem in many cities.

Jacobsen: YDo you see a less explicit version of that transformation with the shift toward an AI-driven technological environment? Are there specific changes that are just as vast, particularly with the loss of entire industries?

Wortel-London: As a professor, I see my students using ChatGPT, but I won’t name names.

Jacobsen: But we know who you are.

Wortel-London: [Laughing] Yes, we know. Students are using AI in new and interesting ways. The danger of automation—a serious concern for factory workers in the 1950s and 60s—is now equally a threat to many creative or finance-based professions.

Entire fields like law, academia, tech, and healthcare are seeing how AI development could lead to job displacement in cities. This presents a real risk for towns that have bet heavily on white-collar, professional-based economies to carry them forward. That strategy has been in place at least since the 1950s.

Before that, in the 1920s, many cities had placed their bets on real estate growth. The Great Depression showed that this alone was not a stable foundation. So after World War II, cities began to think: we’ll still need people doing business together. We need to find the most profitable use of dense urban space.

They believed that use was white-collar work—an early version of today’s “creative economy” arguments. They thought bringing people together in cities would allow them to make faster and better decisions.

But honestly, white-collar work today could be what industrial work was in the 1950s—an economy based on specific technologies or spatial arrangements that are not necessarily permanent. Like the internet, AI can diminish the city’s centrality for some kinds of jobs.

You can see that we’re talking right now over Zoom rather than in a coffee shop. But yes, cities face a real challenge in rethinking their economic raison d’être in the age of AI—and the threat of job displacement that may accompany it.

You can make a city efficient, but that does not necessarily mean employing people. And the deeper, more fundamental point we’re getting at is this: not all growth is good.

Jacobsen: Eric Schmidt, the former CEO of Google, recently said that “growth solves all problems in democracy.” That is, metaphorically speaking, a flat-rate philosophical and economic opinion—which seems untenable in its simplicity. What is the more complicated, more accurate statement?

Wortel-London: On an environmental level, you cannot have infinite growth on a finite planet. That is a physical reality. You cannot have unlimited access to anything – there will be diminishing returns. That’s the second law of thermodynamics—there’s eventually a limit to how much energy, or “juice,” you can squeeze from this planet.

But beyond the environmental level, growth always comes with costs.

When cities built highways in the 1950s, they assumed this would be an “efficient,” cost-neutral activity. However, many of those highways bulldozed neighbourhoods and displaced taxpaying residents and businesses, and cities did not fully feel those impacts until the fiscal crises of the 1970s.

Even in the 1950s, people like Jane Jacobs could sense that something was deeply wrong with these projects. But the powers that be were saying, “Well, this is growth, this is efficiency. Growth and efficiency are good.” In their view, the only solution to the problems caused by growth was more growth.

When we examine the history of development strategies, we see that the logic remains the same: “Growth will pay for itself.” But in truth,  the costs of growth—whether pollution, destructive infrastructure, or unaffordable housing—often hit vulnerable urban populations first. Only later do those costs hit the city as a whole. But when they do, cities often find themselves in the position of their poorest neighborhoods: bankrupt. 

Jacobsen: So these are the so-called economic externalities. However, the term externalities almost minimizes the issue. It is economical to use the trash bin icon on your desktop. 

Wortel-London: It is like a deus ex machina—a way of pretending it will all sort itself out in the end. 

Jacobsen: The assumption that “growth solves all problems” essentially says there is only one way to grow the economy. But if we move away from the question “Should we have growth or not?” and instead ask what kind of growth we want, we will have a far more interesting and productive conversation. What do we want growth for? What is the best way to pursue it? From 1865 to 1981, shortly thereafter, the Soviet Union collapsed. Francis Fukuyama was publishing grand books about the “end of history.” So, those kinds of ideological notions were very much in the air. Is there a distinction between what prosperity meant from 1865 to 1981 and what it came to mean from 1982 to 2025?

Wortel-London: From the 1980s until the financial crisis in 2008, the urban development strategy in many cities remained consistent. What you saw, especially in cities, was an ongoing effort to attract and retain high-income individuals and firms under the assumption that their wealth would help fund desirable social services. If that led to gentrification, that was often considered just the cost of doing business.

There were shifts in how people defined urban growth—not just through ballparks, stadiums, or convention centers, but through more pedestrian-friendly areas and Jane Jacobs-style mixed-use development. Still, the underlying target remained the same: wealthier residents—what we once called “yuppies”—those who could raise aggregate property values and provide a strong tax base.

Since 2008, however, we’ve begun to see the limits of that model—not only within cities but across the country. On the one hand, a few coastal cities in the U.S.—and cities like London or Paris in Europe—are absorbing most of the economic growth. The regions left behind are responding with political backlash, blaming an economy that rewards a small elite at the expense of everyone else. On the other hand, cities themselves are becoming more unaffordable as elites cluster into them. 

One reaction has been exclusionary nationalism, raising tariffs and pulling the drawbridge. But another approach is to follow the money—asking questions like: Who is buying what in this country? Who has purchasing power? Where is that money going? Is it staying within local communities? Is it supporting small businesses?

AI can help with that. At its best, AI can help us make better decisions about the kinds of economies we want to foster by providing clearer assessments of the costs and benefits of different development strategies. We can use it to analyze whether certain growth types—like stadiums and skyscrapers—actually pay off economically.

Should we invest in a new stadium? Should we give Amazon tax breaks to open a facility in our city? Or should we invest in small businesses, community land trusts, and local ownership models? AI can help us evaluate those options.

But it won’t—as long as AI is used merely to make “business as usual” more efficient. In that context, AI is not intelligent, but dumb: It narrows our imagination and limits our ability to think creatively about how cities can grow differently.

Jacobsen: That’s a subtle but important point. The emphasis is usually on increasing computational power or access to large-scale computing, which was once called “computation.” There’s often a focus on the sophistication of the algorithms because brilliant people are designing them. But when it comes to our use of AI—how it affects our thinking, planning, and imagination—it can be constraining rather than liberating. That part of the conversation is rarely brought up. What about cross-sector collaboration to accelerate innovative, AI-empowered urban ecosystems? I mean collaboration between government, academia, and the tech industry.

Wortel-London: When you say “ecosystem,” are you referring to the natural environment, or are you talking more about policy innovation?

Jacobsen: Policy innovation is needed so that friction in collaboration is minimized. You can touch on some of the previously mentioned externalities.

Wortel-London: Right now, AI has much potential to address some of these challenges. We’re already seeing it in initiatives like participatory budgeting and open data platforms that allow citizens—especially those who may not have traditionally engaged in governance—to participate in decision-making about local initiatives.

To the extent that we can outsource certain governmental activities to civil society, that’s a positive development. However, much of the data involved is still proprietary or exclusive. For example, when data is used in housing, law enforcement, or public contracts, and private entities control it, it hinders collaboration and raises serious privacy concerns. I was listening to a podcast recently. The host talked about sensors capable of detecting “antisocial behaviour.” The question is: Who gets to define that? 

Governments should be setting AI risk thresholds, applying a precautionary principle to determine in advance when a given application crosses the line into a rights violation. Should we allow facial recognition technology in public spaces to be governed solely by bureaucrats or private contractors? We should have a say in this, not just bureaucrats and contractors. 

So, community representation—across all segments of society—needs to be involved from the beginning in these technologies’ design, rollout, and implementation.

Jacobsen: How can things go wrong in the implementation of these technologies? People may present excellent proposals on paper, but things often do not go as planned when they are implemented. What typically causes that breakdown? Is it that the idea was never viable, or does the concept not translate well into practice?

Wortel-London: One issue with some of these AI-related policies might be that the technology is simply not ready yet. Take, for example, circular economy models and the ability to accurately measure whether a city’s waste is being reused or recycled back into productive systems. That approach could work well, but it has not yet been implemented effectively on a wide scale.

However, I would say the greater danger is when the systems work too well, but within very narrow parameters. In those cases, they tend to reproduce the assumptions and biases embedded in them from the start.

If you use algorithms to promote flawed development strategies, you are making them more straightforward to implement and more challenging to reverse. They’ll be rolled out at such speed and scale that contesting them after the fact becomes much harder.

This also applies to bias in law enforcement. If algorithms are used to prescreen individuals or predict where crime is more likely to occur, they may disproportionately concentrate policing in low-income neighbourhoods.

But what about crimes in finance? What about crimes on Wall Street? Will we redefine crime to include forms of harm that rarely make headlines? The risk is that these tools—rather than expanding our vision—may simply reproduce our existing assumptions.

There’s that old saying: Everything looks like a nail when you have a hammer. In the age of AI, it becomes: when you have AI, everything looks like data. But what ultimately makes cities valuable are the stories and experiences of the people living in them—not all of that can be reduced to ones and zeros.

Think about what Jane Jacobs would say about AI. If there’s anyone who could be considered the patron saint of urbanism, it is her. Her vision of the city was that of a street ballet—a complex, spontaneous choreography of human life. I am not sure there is an algorithm or equation for that.

But it would require flesh-and-blood human interaction. The top-down technocracy that Jane Jacobs opposed in the 1960s is not so different from the kind of technocracy we might face now if we hand over governance decisions to AI without serious public oversight.

So, we need to remember past struggles—against displacement, against highways, against destructive development projects—and ask: how might AI repeat those harms if we’re not careful?

Jacobsen: Daniel, thank you so much for your time today. I appreciate your expertise—and it was a pleasure to meet you.

Wortel-London: Yes, thank you so much, Scott. I hope some of this was useful.

Last updated May 3, 2025. These terms govern all In Sight Publishing content—past, present, and future—and supersede any prior notices. In Sight Publishing by Scott Douglas Jacobsen is licensed under a Creative Commons BY‑NC‑ND 4.0; © In Sight Publishing by Scott Douglas Jacobsen 2012–Present. All trademarks, performances, databases & branding are owned by their rights holders; no use without permission. Unauthorized copying, modification, framing or public communication is prohibited. External links are not endorsed. Cookies & tracking require consent, and data processing complies with PIPEDA & GDPR; no data from children < 13 (COPPA). Content meets WCAG 2.1 AA under the Accessible Canada Act & is preserved in open archival formats with backups. Excerpts & links require full credit & hyperlink; limited quoting under fair-dealing & fair-use. All content is informational; no liability for errors or omissions: Feedback welcome, and verified errors corrected promptly. For permissions or DMCA notices, email: scott.jacobsen2025@gmail.com. Site use is governed by BC laws; content is “as‑is,” liability limited, users indemnify us; moral, performers’ & database sui generis rights reserved.