Photo: Older men talking together on a street in the old town of Bhaktapur, Nepal. Credit: Roman Saienko via Pexels.

No one announced it. There was no day when the city stopped being human and started being run by machines. Instead, the change arrived one convenience at a time: a self-checkout instead of a cashier, an app instead of a trip to the market, a chatbot instead of a clerk, a route chosen by an algorithm instead of by habit or curiosity. Each step made life a little easier. Together, they are quietly changing what it feels like to live in a city.

There is nothing wrong with artificial intelligence itself. It can save time, cut waste and make services fairer and faster. The question this essay asks is a different one: as cities are increasingly designed, managed and optimised by machines, will human emotion still matter? And if we lose the humanised city slowly, will we even notice, and will we be able to get it back?

A change too slow to notice

Ecologists have a name for this kind of slow loss. In 1995 the fisheries scientist Daniel Pauly described the “shifting baseline syndrome”: each generation of scientists takes the fish stocks it first encounters as the normal baseline, so each new generation starts from an already depleted ocean and gradually accepts the disappearance of species as ordinary [1].

Cities may be going through something similar. When Amazon opened its first cashier-free Go store to the public in Seattle in January 2018, cameras along the shelves tracked what shoppers picked up and billed them automatically as they walked out, with no checkout at all [2]. At the time it felt like science fiction. Today, contact-free payment, app-based ordering and automated service desks are simply how many cities work. A child who grows up ordering food on a phone, navigating by a map app and paying without speaking to anyone does not experience these things as a loss. What older residents remember, the chat at the corner shop, the stranger who gave directions, the long wait that turned into a conversation, does not disappear in a single moment. It fades, and the baseline moves with it.

The sociologist Sherry Turkle captured the emotional side of this shift in the subtitle of her book Alone Together: we increasingly expect more from technology and less from each other [3]. Shoshana Zuboff has described a deeper economic logic behind it, in which everyday human experience becomes raw material for behavioural data [4]. Even work is changing. The International Labour Organization reports that digital labour platforms grew five-fold worldwide in a decade, and that algorithms are increasingly replacing humans in allocating, evaluating and monitoring work [5].

The city as a place of encounters

Cities have always been more than infrastructure. In The Death and Life of Great American Cities (1961), Jane Jacobs described the everyday life of a busy street as a kind of “sidewalk ballet”, in which countless small, unplanned contacts between neighbours, shopkeepers and strangers build trust and safety over time [6]. The sociologist Ray Oldenburg argued that cafés, barbershops, bookshops and other “third places”, distinct from home and work, are where informal public life and community are made [7]. The Danish architect Jan Gehl showed that the quality of streets and squares shapes how much optional and social activity takes place in them: good public space invites people to stay, meet and linger [8].

Emotion is part of this urban fabric, and it can even be measured. In his Bio Mapping project, the artist Christian Nold gave participants a device combining GPS with a skin response sensor, sampling their physiological arousal every four seconds as they walked, to produce “emotion maps” of neighbourhoods [11]. The Mappiness study, which collected more than a million mood reports from over 20,000 people in the UK, found that people were significantly happier outdoors in green or natural settings than in urban environments [10]. Writers such as Charles Montgomery have argued that city design is, in the end, a question of happiness as much as efficiency [9].

These encounters and feelings are inefficient by design. They are slow, unpredictable and hard to measure. That is exactly what makes them valuable, and exactly what makes them vulnerable in a city increasingly organised around speed, convenience and data.

What the numbers already show

Worries about a less social life are not new. Two decades ago, Robert Putnam’s Bowling Alone documented a long decline in American social and civic life [12]. The trend has continued. According to the American Time Use Survey, Americans spent an average of 34 minutes a day in 2023 visiting friends or attending and hosting social events, down from 43 minutes a decade earlier and 47 minutes twenty years earlier. Analyses of the same survey suggest that face-to-face time with others fell by around 30 per cent between 2003 and 2022, and by more than 45 per cent among teenagers [13].

Globally, the World Health Organization’s Commission on Social Connection reported in 2025 that one in six people worldwide is affected by loneliness, and that loneliness is linked to more than 871,000 deaths every year. Between 17 and 21 per cent of people aged 13 to 29 reported feeling lonely, with the highest rates among teenagers [14]. This matters for health, not only for happiness. A landmark meta-analysis of 148 studies and more than 300,000 participants found a 50 per cent greater likelihood of survival for people with stronger social relationships [15], and the US Surgeon General warned in 2023 that lacking connection can raise the risk of premature death to levels comparable to smoking daily, with about one in two American adults reporting loneliness [16].

It would be wrong to blame all of this on AI or digital technology. Research cited in the same reporting suggests that the decline in face-to-face time began well before social media [13], and loneliness has many causes, from long working hours to changing family structures. But it would also be naive to assume that a city designed to reduce human contact at every step will leave our emotional lives untouched.

Optimised for efficiency, not for feeling

AI works by measuring, predicting and optimising. It is very good at what can be counted: travel time, energy use, queue length, cost per transaction. Emotion is much harder to count. A conversation at a bus stop, a sense of belonging in a neighbourhood market or the comfort of a familiar face rarely appear in a dataset, so they rarely appear in the objective a system is asked to improve.

The result is not a dramatic dystopia but a series of reasonable decisions. A navigation app sends traffic down the fastest road, not the most pleasant one. A delivery platform removes the need to visit the shop. An automated counter removes the need to speak to anyone. Each decision is efficient. Together, they can hollow out the small human moments that make a city feel like home.

UN-Habitat has warned that AI is appearing in cities at an unprecedented rate, often with significant risks and little oversight, and that using it without proper governance or attention to human rights can have negative, even catastrophic, effects [17]. People also push back when they sense the balance tipping. In Toronto, Sidewalk Labs, a company owned by Google’s parent Alphabet, proposed a technology-led waterfront neighbourhood at Quayside. The plan faced sustained objections over privacy because the company intended to collect data as part of its smart city vision, and in May 2020 Sidewalk Labs abandoned the project, citing the economic impact of the COVID-19 pandemic [18]. Public unease is widespread: in 2023, 52 per cent of Americans said they felt more concerned than excited about the growing use of AI in daily life, up from 38 per cent less than a year earlier [21].

Not anti-AI, but pro-human

The answer is not to reject AI. Used well, it can make cities more humane: translating for newcomers, helping people with disabilities move around, spotting where residents feel unsafe, or freeing staff from routine tasks so they have more time for people. The difference lies in what the technology is asked to serve. The same tools can be designed to replace human contact or to make room for it.

Some cities are already setting limits and asking for transparency. In September 2020, Amsterdam and Helsinki became the first cities to launch public AI registers, explaining which algorithms they use in public services, from chatbots to parking control, and why [20]. The European Union’s AI Act, which came into force on 1 August 2024, bans real-time remote biometric identification in public spaces, with narrow exceptions for law enforcement [19]. These are signs that the human dimension of the city can be defended, if we choose to defend it.

Keeping the city human

Keeping the city human will not happen by accident. It requires deliberate choices by planners, designers, technology companies and citizens. Cities can protect and invest in third places, from markets and libraries to parks and tea stalls. They can insist that essential services keep a human option. They can measure social connection and wellbeing, not only speed and cost, when they evaluate new technology. And they can give residents a real voice in how data about their lives is collected and used.

Above all, we need to notice the change while it is still happening. The danger of a slow shift is that, by the time it becomes obvious, the baseline has already moved and there is no one left who remembers what was lost. Pauly’s fisheries offer a warning: some things are far easier to keep than to restore.

Human emotion is not an inefficiency to be engineered out of the city. It is the reason cities exist. The machines can help us run our cities better, but only we can decide that they should remain places for people to feel, to meet and to belong.

Connect with the author, Asesh Sarkar, on LinkedIn, and follow Urban Frontier Insights on LinkedIn.

References