The case for… cities that aren't dystopian surveillance states

Cory Doctorow (@doctorow)

January 17, 2020

Imagine your smartphone knew everything about the city ‐ but the city didn't know anything about you. Wouldn't that be truly "smart"?


Smart city is one of those science fiction phrases seemingly designed to make you uneasy, like "neuromarketing" or "pre-crime". It's impossible to be alive in this decade and not find something unsettling in the idea of our cities becoming "smart".

It's not hard to see why: "smart" has become code for "terrible". A "smart speaker" is a speaker that eavesdrops on you and leaks all your conversations to distant subcontractors for giant tech companies. "Smart watches" spy on your movements and sell them to data-brokers for ad-targeting. "Smart TVs" watch you as you watch them and sell your viewing habits to brokers.

Why isn't it creepy for you to know when the next bus is due, but it is creepy for the bus company to know that you're waiting for a bus?

Smart cities are studded with sensors that monitor what's going on with people, vehicles, and infrastructure; and use actuators to change things based on the resultant data.

Put that way, it's hard to imagine a city that's not "smart". When you call 999, you are acting as a sensor. The fire brigade comes roaring to the rescue in a fire engine ‐ that is, a giant, high-speed, actuating robot. Transit systems are all sensors ("Is there a train ahead of me still at the platform?") and actuators ("Hit the brakes!"), and they've been steadily exposing more and more of the data they generate to potential riders, so you can text a number or use an app or check a lighted signboard to find the wait time until the next vehicle.

All this raises an interesting question: why isn't it creepy for you to know when the next bus is due, but it is creepy for the bus company to know that you're waiting for a bus?

It all comes down to whether you are a sensor ‐ or a thing to be sensed. In the "internet of things," we're promised technology that will allow us to project our will on to our surroundings, changing our lighting or unlocking our doors or adjusting our thermostats from anywhere in the world. But anyone who's used these technologies for more than a few minutes quickly starts to suspect that they are also a thing, just another thing to be sensed and acted upon from a distance, generally by unaccountable algorithms seeking to corral us into altering our conduct to maximise returns to their manufacturers' shareholders.

As with cities, homes were sensing and actuating long before the "internet of things" emerged. Thermostats, light switches, humidifiers, combi boilers… our homes are stuffed full of automated tools that no one thinks to call "smart," largely because they aren't terrible enough to earn the appellation.

Instead, these were oriented around serving us, rather than observing or controlling us (with rare exceptions, such as electricity and gas meters, which were designed on the assumption that they were going into hostile territory and that we couldn't be trusted not to tamper with them). In your home, you are not a thing, you are a person, and the things around you exist for your comfort and benefit, not the other way around.

Shouldn't it be that way in our cities?

There's nothing wrong ‐ or new ‐ in the idea that we should sense what's happening in our built environments and alter how our systems perform to respond to those sensors' observations. There's nothing objectionable about adding more trains when the system is busy, or recording accurate usage data to inform our urban planning debates. The problem is that the smart city, as presently conceived, is a largely privatised affair designed as a public-private partnership to extract as much value as possible from its residents while providing the instrumentation and infrastructure to control any civil unrest that such an arrangement might provoke. Far from treating residents as first-class users of smart infrastructure, they are treated as something between gut flora and pathogen, an inchoate mass of troublesome specks to be nudged into deterministic, convenient-to-manage patterns.

It needn't be this way. As is so often the case with technology, the most important consideration isn't what the technology does: it's who the technology does it to, and who it does it for. The sizzle reel for a smart city always involves a cut to the control room, where the wise, cool-headed technocrats get a god's-eye view over the city they've instrumented and kitted out with electronic ways of reaching into the world and rearranging its furniture.

It's a safe bet that the people who make those videos imagine themselves as one of the controllers watching the monitors ‐ not as one of the plebs whose movements are being fed to the cameras that feed the monitors. It's a safe bet that most of us would like that kind of god's-eye view into our cities, and with a little tweaking, we could have it.

If we decide to treat people as sensors, and not as things to be sensed ‐ if we observe Kant's injunction that humans should be "treated as an end in themselves and not as a means to something else" ‐ then we can modify the smart city to gather information about the things and share that information with the people.

Imagine a human-centred smart city that knows everything it can about things. It knows how many seats are free on every bus, it knows how busy every road is, it knows where there are short-hire bikes available and where there are potholes. It knows how much footfall every metre of pavement receives, and which public loos are busiest.

What it doesn't know is anything about individuals in the city. It knows about things, not people. All of that data is tremendously useful to the city's planners and administrators, of course, as a way of planning and optimising services, infrastructure and future building.

But it could also be useful to the people in the city.

While we're imagining a city that is instrumented to measure things but not people, try imagining a mobile device that gathers data about its user, but doesn't ever share that data with anyone, ever. Its backups are encrypted to a passphrase that only the user knows, and it jealously guards the data about its use from the vendors who supply its apps. This device knows everywhere you go, it knows what you buy, it knows whom you talk to and how long and maybe even what about. In other words, it is extremely similar to the device you're carrying around right now ‐ with the vital difference that it keeps what it learns about you private.

Now, equipped with your device, you are prepared to be a sensor, rather than a thing to be sensed. As you move around your smart city, the things around you stream data about their capabilities, limitations, prices, uses and nature. Want to find a loo? Your device not only knows which ones are free, but also what time you habitually pee, and whether or not you've been drinking a lot of water and might need one. Want a free seat on a bus? Likewise, the device will tell you where there is one free. When you stand at a bus-stop, your presence, but not your identity, is registered, so that the transit system can adjust the vehicles and routes.

All of this is simply broadcast to all the devices in the vicinity, and your device can "tune in" to a stream of data simply by plucking it out of the electromagnetic spectrum, without activating a connection to the server that would leave a record of what you took an interest in.

If you want to page a minibus ‐ something like an Uber Pool, but run by the city, licensed, safe, paying a living wage and not mining your data ‐ you can summon one, and yes, this exposes your identity so that the driver can find you.

This is an example of how a smart city could work: a place through which you move in relative anonymity, identified only when needed, and under conditions that allow for significant controls over what can be done with your data.

Such a city depends on a responsive, legitimate government, and on devices that are open and transparent, freely auditable and secured through widespread scrutiny of their inner workings. It is a city and a technology and a government oriented around its people, designed to treat people "as an end in themselves and not as a means to something else".

If it sounds utopian, it's only because of how far we have come from the idea of a city being designed to serve its demos, rather than its lordly masters. We must recover that idea. As a professional cyberpunk dystopian writer, I'm here to tell you that our ideas were intended as warnings, not suggestions.