It’s time to panic about who’s watching us

Farhad Manjoo
Published : 14 April 2019, 09:17 PM
Updated : 14 April 2019, 09:17 PM

"Privacy" is something we all seem to want. You get mad when your privacy is invaded or in some way mishandled; when Equifax leaks your credit info, when your sexts show up on Reddit, when your psychiatrist's office gets hacked.

Privacy is also something many people are happy to trade away at a moment's notice, for the slightest reward: your address to win this car, your purchase history for discounts on peanut butter, your life for Facebook.

This sounds like a contradiction, but it's more of a resignation. People claim to want privacy, companies claim to provide it, but neither side is being entirely transparent or honest about the terms of the trade. When something goes wrong, we all get very angry for a minute and then go back to business as usual. Secretly, we might already have made peace with the idea that no one has any privacy online.

It's time to stop shrugging about privacy.

The next decade will be marked by the rise of technologies that are deeply dependent on the most intimate information about our lives. I'm talking about things like artificial intelligence, wearable sensors, genetic sequencing, drones and tiny cameras everywhere. Then there's the emerging dynamic of the tech industry, in which a small number of all-powerful giants are becoming central collectors of all digital information. These facts point us toward doom. Soon, a handful of companies will enjoy almost total information awareness over just about everything that every one of us is doing at all times.

Let's play a game I like to call "This Gadget's Worst Nightmare." The game goes like this: Pick some newish tech thing. Then spin the story out in your head. Imagine all of the ways the technology might be misused, and ask yourself if our laws and politicians are prepared to handle the dystopian nightmare that emerges. I'm going to warn you: The game gets very scary very quickly.

Consider home surveillance cameras. In particular, think about doorbell cams, which appeared on the scene a few years ago and became an instant hit. They're handy — when people ring your bell, you can see their faces and chat with them. Today, two of the most popular home cameras brands are owned by tech giants: Amazon makes Ring, and Google owns Nest.

Privacy-wise, these cameras might initially seem like no big deal — just a new take on the old video intercoms you'd see on jewellery stores in '80s heist movies. But if you consider their dystopian possibilities, doorbell cameras raise many profound social, moral, economic and even political questions.

First, they are motion-triggered cameras that are connected to the cloud. This means that by installing a doorbell cam, you are recording an image of everyone who comes to your door and sending pictures of that person's face to unknown servers far away. This might sound OK — after all, it's your house, and you have a right to film people — but remember that your property can also be someone else's workplace. Delivery people, for instance, might have to come to your door every day. Without obtaining any permission, you're sending a daily dossier of a UPS driver's face — and location at a certain time of day — to some internet company.

Is that OK?

While you figure that out, here are some other complications. Manually screening every face that appears at your door is a trudge. Luckily, software can do it for you. Through face recognition, doorbell cameras can be taught that certain people are friendly and others are not. This could be convenient: You're travelling on business and you want to be alerted when your spouse gets home. But it could just as easily turn nightmarish: A domestic abuser wants to be alerted when his spouse gets home. Indeed, "smart home" gadgets are already becoming a key weapon of domestic abuse.

Now imagine what might happen when the doorbell camera finds someone it doesn't recognise. Ring, the doorbell camera startup owned by Amazon, offered some creepy ideas in a patent application it filed last year. The company described a way to stitch together a composite of a face that might have been captured by several different cameras. The face could then be remembered and shared by neighbours, and checked against a "database of suspicious persons" — either databases created by the neighbours themselves, or perhaps law-enforcement or immigration databases.

This sounds like the far-off future, but much of it is here today. Ring already offers a neighbourhood social network that lets users share pictures of "suspicious" people. Note also that neighbourhood social networks are known hotbeds of racial profiling, and that facial recognition software is shot through with gender and racial bias. On top of all that, Ring offers ways for the police to tap into its neighbourhood networks: As The Intercept reported, cops can look up neighbours' comments and even request their video footage.

Do we want a semipublic or even fully public, searchable database of everyone's "suspicious" doorbell pictures? Is that OK?

But it gets worse. Doorbell camera companies say they don't spy on your videos. But they do not forswear using other information gleaned from the smart home — like a log of times you're likely to be home and not — to make predictions about you, to develop new products and business ideas, and to better market to you. And it's not just the doorbell camera manufacturers you have to worry about. Last year, Congress repealed privacy rules meant to prevent broadband companies from analysing your private information; as a result, your internet service provider might well be able to glean information about your comings and goings from your doorbell camera — and perhaps to sell it.

Think about all the economic, social and even political power we're handing over here. Amazon and Google already know loads about you. By blindly buying into these tiny conveniences, we are giving them eyes on every other door in America. Is that the kind of information we want private corporations to have?

And someday the decision may not even be voluntary. One life insurance company is now requiring the people it covers to wear activity trackers. What if your homeowners insurance company gives you a significant discount for installing home cameras? What if the doorbell cameras take off, but not so much because people love them but because they don't hate them enough to say no to what looks like a convenient bargain?

Now play this game for other technology: drones, ancestry databases, self-driving cars …

To examine how carelessly we're thinking about privacy as we head into this digital future is to be confronted with a stark truth: We are building a surveillance state no less totalitarian than the one the Chinese are rigging up. But what China is doing through government, we are doing through corporations and consumer products, in the absence of any real regulation that recognises the stakes at hand.

© 2019 New York Times News Service