September 24, 2026, (Inside AI) — A new book from MIT researchers argues that visual artificial intelligence can transform urban planning by turning millions of street-level images into actionable data, but warns that the same technology risks eroding privacy and reinforcing social biases.
"How AI Sees the City: Urban Visual Intelligence," published this month by Routledge, examines how computer vision systems can analyze traffic patterns, emissions, green space, and even interior design trends at a scale never before possible. The authors are Fabio Duarte, Martina Mazzarello, Carlo Ratti, and Fan Zhang.
The book arrives as cities worldwide grapple with how to deploy camera networks responsibly. London operates roughly 210 cameras per square mile, while Shanghai has over 5,000. In the United States, debates over traffic camera usage intensified this year, with privacy advocates warning that constant monitoring could chill public life.
"We can treat these digital images as data and quantify features of the city," says Fábio Duarte, a principal research scientist and associate director of the MIT Senseable City Lab. "With computer vision techniques, each image is a dataset." Still, he adds, "We have to be careful about it."
Read: Pune Deploys 10,000 Police and AI Surveillance for Ganesh Immersion
The researchers position visual AI as the latest chapter in a long tradition of visual urban analysis. They cite Kevin Lynch, whose 1960 book "The Image of the City" shaped modern urban design, and William H. Whyte, who pioneered the study of public spaces. Lynch worked with pen and paper. Today's tools can process millions of images simultaneously.
"Kevin Lynch at MIT was only using paper and pen," Duarte says. "We can now scale up what he was doing, with visual AI, while also looking at many different dimension of cities."
The Senseable City Lab has already demonstrated the potential. In a study earlier this year, researchers used machine learning to identify vehicle types from 331 traffic cameras in New York City and estimate emissions from each automobile. The method could enable precise, city-wide emissions monitoring. Other applications include analyzing intersection safety, park usage, and street-level greenery.
"The real promise of visual AI is not simply that computers can look at millions of images," Zhang says. "It is that we can connect what is visible in those images -- streets, buildings, greenery, traffic, public space -- with larger questions about how cities function and how people experience them."
One recent Senseable City study analyzed 400,000 Airbnb listings worldwide and found that interior design styles are not becoming globally homogeneous, contrary to some claims. The research suggests that visual AI can reveal geographic differences that might otherwise go unnoticed.
But the authors devote significant attention to the pitfalls. They write that the benefits of intensive video recording "must be weighed against the significant erosion of personal freedom and the potential for abuse inherent in a system of constant monitoring."
Read: California Governor Issues Executive Order on AI Safety
AI systems can also perpetuate social biases. If models are trained primarily on majority population groups, they may not evaluate minority groups accurately. The result can be data that reinforces existing perceptions rather than reflecting underlying realities.
"We need to teach AI to see, and depending on how you teach it, it will see what what is embedded in the culture," Duarte says. "AI is not neutral."
Mazzarello offers a counterpoint: human perception is also shaped by cultural bias. "Our eyes are not neutral, either," she says. "Every tool has to be guided in the right way, and trained in the best way."
The book has drawn praise from other scholars. Michael Batty of University College London called it a "fascinating book" that "shows how we are beginning to interpret the world of urban design, suggesting ways in which we might improve design using urban analytics, AI and large language models."
Despite the risks, the authors remain optimistic. They argue that visual AI, used critically and creatively, can help cities become more livable, equitable, and sustainable. As they conclude in the book, "We should explore this wisely, critically, and creatively."
