Horizons from PBS News

I'm William Brangham, and this is "Horizons."

We are all being watched in ways we never imagined.

Vehicle tracking, facial recognition technology, data mining.

What does all this digital surveillance mean for the right to privacy?

And as a backlash over this monitoring grows, what can you do to protect yourself?

The ethics of surveillance coming up next.

Narrator: Support for "Horizons" has been provided by Steve and Marilyn Kerman and the Gordon and Betty Moore Foundation.

Additional support is provided by Friends of the News Hour.

This program was made possible by contributions to your PBS station from viewers like you.

Thank you.

From the David M. Rubenstein Studio at WETA in Washington, here is William Brangham.

Welcome to "Horizons."

Do you have a sense of how often you're being tracked, how your movements are being recorded, how what you type or speak into your phone or your computer is being captured and cataloged?

We live in an era of remarkable connectivity, thanks in large part to the Internet and to all the devices that we own that link to it.

But we're also living in an age of unprecedented surveillance, aided not only by those many devices, but also by the proliferation of video cameras all over our society.

Couple all that with a range of companies that harvest and sell so much of our personal data and the idea of a right to privacy seems to have been hijacked.

The nationwide deployment of Flock cameras, which use AI to capture and track the license plates of passing cars, is just the latest example of a growing protest movement.

The allegation being that all those eyes in the sky violate people's basic rights.

So we wanted to look at the current state of this technology, how information about us is being monitored, captured, sold and used.

And for that, we are joined by two people who have wrestled with this technological dragnet for years.

Byron Tau is a Pulitzer Prize winning investigative reporter with ProPublica.

He's the author of the 2024 book, "Means of Control, how the hidden alliance "of tech and government "is creating a new American surveillance state."

And Calli Schroeder is senior counsel and director of the AI and Human Rights Program at EPIC, the Electronic Privacy Information Center, which is a research and advocacy group that's focused on privacy and freedom of expression.

Welcome to you both.

Thank you so much for being here.

Byron, I would love to start with you first.

You heard my introduction, this depiction of a sort of web of surveillance and all of the different ways that we are being looked at.

And I can already hear some viewers thinking, "You're being hyperbolic, William, "you're making up an overly negative view."

I know you don't think necessarily this is hyperbole, but imagine that our viewers might think that.

Give us a little bit of a primer on what you would argue why that point is actually true.

Yeah, and thanks for having me.

It's a great question.

So I think it's important to put this into context, that never before in human history has so much information been in both private and government hands, not during eras of totalitarian government.

At no point has there ever been so much information out there.

And that includes everything from the license plate readers you mentioned to the data that our phone emits, to the basic information that our governments collect, to everything that you give to corporations, whether that's your emails, your very private search history, your chats with chatbots, all that information is in someone's hands.

And it's available.

It's available to public safety entities.

It's available to government agencies.

It's available to the military.

It's available to the intelligence community.

And so we as a society have historically relied on limiting government access to information in order to preserve liberty.

And with so much information floating out there about all of us, both in private and government hands, I think it calls into question some of the basic assumptions that have gone into building this country's political system.

Calli, you heard what Byron has to say there.

What would you add to that?

I would add that I think part of the reason people don't realize how much surveillance they're constantly under is because it's often couched in friendlier terms or more appealing terms.

So they don't say that, "We're watching everywhere you go online."

They say, "We're tracking information "so we can personalize things for you "so we can do tailored advertising "so that we can optimize our services."

It's all framed in a way that makes it seem like it's actually for you when what's happening is private companies and governments and data brokers and all of these other entities that have built up this whole ecosystem of information are actually taking information from you for their own monetary and other gain.

So surveillance is happening everywhere.

We just aren't always trained to recognize it.

Brangham: You know, Byron, as Calli is talking, I'm reminded of... in George Orwell's "1984," that every person is issued what they call the telescreen, which is basically this television in your in your home that broadcasts government propaganda.

But it also looks at you and watches you and reports back if you're acting inappropriately.

That, in the novel, everyone knew was a hostile force that they needed to try to avoid.

Here, as Calli is describing, a lot of these devices we have welcomed into our lives and put them on our wrists and our pockets and on our desks.

Like that, it seems like we have invited a lot of this technology as opposed to it being imposed upon us.

I think that's right.

You said it really well, that that these things have been put in our home for our entertainment, for our convenience, to help us communicate.

But each one of these devices that we put in our home or these services that we sign up for, they do have surveillance potential.

And people don't like to think about that, right?

That the Echo Dot that they put in their house has a microphone.

It has the capability to be hacked.

A government could go get a search warrant and listen in in some instances.

The ring doorbell that you put on the front of your house for security and to make sure your packages don't get stolen, that has surveillance potential.

Police request that footage all the time.

And so, as you say, we've welcomed many of these devices in without thinking through the consequences as a society of, do we want that kind of third party access in our home?

And what are the rules that we as a society want to write about where to limit these powers of both corporations and governments?

That's right.

Calli, you mentioned in an earlier conversation... Byron was just talking about the ring doorbell.

They ran this this infamous ad during the Super Bowl last year.

It was ostensibly about, ok, let's say you've lost your beloved dog.

Ring doorbells got a feature that can help you figure out how to find your dog in your neighborhood.

Let's take a little bit of a look at that.

Narrator: One post of a dog's photo in the Ring app starts outdoor cameras looking for a match.

Search Party from Ring uses AI to help families find lost dogs.

Since launch, more than a dog a day has been reunited with their family.

Siminoff: Be a hero in your neighborhood with Search Party, available to everyone for free right now.

Now, that ad absolutely blew up in Ring's face.

I mean, you can see how very simply this beloved idea of finding a lost dog could be used for very different purposes.

Yeah, it was really fascinating to me.

So I've worked in the privacy space for a decade now, and it was really fascinating to me seeing so much immediate, coordinated, published public pushback to this ad because it was framed in such a positive and benign way.

And frequently, companies have been able to successfully launch these surveillance tools by framing them appropriately.

And in this case, it absolutely did not work.

There was massive public pushback.

Immediately, people clocked that this could be used to track bystanders, people just walking around a neighborhood.

Suddenly, they're being tracked all over the place by what are essentially cameras everyone has installed on their homes.

And that reaction was both interesting and, as someone that's very pro-privacy, fairly heartening that people were paying that much attention.

Brangham: I mean, Calli clearly doesn't want people to find their lost dogs out in the world.

So she's a villain in this case, too.

Byron, but the point is also, yes, we have brought some of these devices into our homes, but the other the technology that I mentioned, Flock cameras, those are not put up by individuals.

Those have been put up usually by local government.

For people who are not familiar with what those are, what are Flock cameras?

Where are they being deployed?

What's the controversy around them?

Sure.

So Flock is one vendor of a technology called a license plate reader, which basically is a fixed camera.

And actually, sometimes they're mobile cameras and they're attached to police cars or garbage trucks or something like that.

And they essentially record the location of where they saw a plate in the world.

And Flock is just one vendor and it's been in the news a lot, but there are many other vendors of this technology.

And over the past decade plus, many states, counties, cities have put these cameras up and they're all being networked together.

They're all searchable.

And essentially what they do is they create a record of where automobiles go in the world.

And, you know, an automobile is a proxy for you.

We Americans love our cars.

That's how we get around 90 plus percent of the time.

And so I think there's a growing sense among Americans that everywhere they go, they're being monitored.

And I think that's part of what you see in some of this backlash against this technology.

Calli, my colleague, Paul Solman, did a story about the controversy around the Flock cameras, and he spoke to a spokesperson for Flock and they argued this technology is wonderful at solving crimes.

I think the spokesperson claimed that something like 800,000 to a million crimes per year were solved because a Flock camera was able to catch a perpetrator moving from one place to another or driving or in the act.

What do you make of that argument?

Because that is the argument that cities and law enforcement argue is why we need these cameras, because it stops crime and it catches criminals.

I think it's a disingenuous argument for quite a few reasons.

Number one, it implies this is the only way to solve crimes.

It's certainly not.

Police have absolutely solved many, many crimes before automated license plate readers were all over the place.

It also implies that every time Flock flags someone's location or links them to a potential crime, it's correct, which is certainly not true.

I'm from Colorado.

There is an active case going on where a woman had a police officer come to her door and say, "We know that you've been doing package theft "in this area because we've been able "to track your car going into this neighborhood "at this time of day "and exiting at this time of day."

They did not actually tie her to the specific location.

There was absolutely no evidence that she had had anything to do with these things, but a cop came to her door and told her, "You're going to be arrested "because we have this evidence against you."

That's a misuse of the technology and in some ways can actually make law enforcement lazier at their jobs, where they're having Flock make these automated decisions for them or pull this information together.

And it's not correct.

And human judgment would be able to see the gaps in that case.

The final thing I'll note is that while a big push for these companies is that this can help solve or prevent crime, I want to really highlight that any argument that more surveillance means more safety, as in less crime, as in it will prevent crime, is not true.

Cameras everywhere doesn't stop crime from happening.

It just makes it easier to observe.

So the question here is not a matter of can we stop crime with surveillance, it's how much of your civil liberties are you willing to give up so police have a better view of crimes that are committed?

Tau: And I think... If I just could add to that, I think a lot of the ways that these technologies, particularly license plate readers, are sold to the public is exactly what you just said.

That we're going to go out there and we're going to identify known getaway cars or vehicles associated with kidnapping children.

And police do use them for that.

But there's also other uses that are a little bit more secretive and a little bit more nefarious.

And that includes things like predictive policing.

And that includes things like... Brangham: Predictive policing.

Yes.

So federal agencies... I did a story on this when I was at the Associated Press last year, but federal agencies are taking license plate reader data and they're looking at the patterns of people moving around in the world.

Not you are a known criminal, but you drive this particular route and that pattern of your travel, going from a big city to the border region and back in the same day, is inherently suspicious to us.

Brangham: That looks like what a human trafficker... Tau: Or a drug smuggler or just something that we want to investigate a little bit further.

And so we're going to find a way to flag your license plate to a local sheriff.

They're going to find a reason to pull you over.

Were you going two miles over the speed limit?

Do you have a busted mirror?

And we might even search your car.

We might detain you.

We might, you know, ask you to consent to opening your phone or answer all sorts of invasive questions.

Brangham: Are there documented cases of this kind of thing happening?

Tau: There are.

Yes.

There's a lawsuit in Texas right now over this.

And there's a Border Patrol intelligence program that looks at the movement of people around the United States and says and singles them out for suspicion based on where they come from, where they go and facts about them.

Calli, Byron's talking about here, the inference that can be made from certain behaviors, certain... certain data points that one of these devices gives to law enforcement or to a corporation that they can then stitch together a better picture of who you are.

Again, sometimes this is pitched as, "Look, we want to know that you buy hiking boots "and that you drive a Subaru "because we might think you're going to be interested "in a hiking vacation package from some other company."

But there are other instances where that's not the case.

Can you sort of explain a little bit how inferences like this work in this world?

Yeah, absolutely.

So part of my role in tracking AI developments at Epic has been looking at how AI develops and how the risks that we're seeing from this technology are different or the same from risks that we've seen across different kinds of privacy harms.

And in my opinion, the big difference is in the scope and scale of the harm there.

So as you mentioned, tons and tons of companies, private companies, companies with government contracts, they are collecting information on people all the time.

There are tons of data points.

There are whole businesses called data brokers where their entire job is collecting as much information on you as possible from every single page you go to online to any real time information, location tracking they can get to what stores you go to, to what people you talk with.

And they build these files, they sell them to advertisers, and that is used to do targeted advertising.

But they're also sold to other groups.

So, for example, the government could buy this data broker file on you and use it for analysis without a warrant because it's available for purchase.

And AI is able to go through all of this information, maybe combine these massive files from multiple data brokers, maybe pull all public information it can scrape from online.

And it's able to go through all of this information and make inferences from it.

So maybe, just as an example, maybe, if I was a queer person, maybe I've never said online that I'm queer, maybe that is private for me and I've never shared that.

But if it's able to go through every picture that other people have posted of me, if it's able to track my location, if it's able to track who I talk to most often, what kind of organizations I look at, what kind of articles I read, it may be able to infer my sexuality.

And then that becomes part of a file somewhere.

That's now information that's out in the world that I did not put out there.

So the risk of inferences and pattern recognition and the amount of data that AI is able to process to pull these inferences, that's one of the major risks that we're looking at.

It's so such a dark and tangled connections you're making there.

Byron, what is the law about this?

Is this illegal?

All of the things that you are describing, are we protected by this?

Does the so-called prohibition against unreasonable search and seizure apply here?

Is there anything that makes this kind of thing unacceptable?

Well, we should probably direct the question at the lawyer.

But generally speaking, you know, the government has taken the position that, if there's data out there that's either publicly available and you can just go and look at it, or if there's data out there in a marketplace that's available to buy, that that isn't a Fourth Amendment violation.

And so that's been an end run around the traditional protections that have been in place since the founding of this country in terms of the access to information that government can get.

Brangham: Meaning that if the... Please, Calli, as the lawyer here.

No, just adding to that, Byron's absolutely correct.

Not... Absolutely not to get too deep into the legal weeds, but this is a development that is referred to often as the third party doctrine, which is if you've given information to a third party, even if it was for a specific purpose.

So even if I did share information, but it was with Microsoft, so I can do my job because I need Microsoft tools to be able to write documents, because I voluntarily shared information with that company, that now can be requested by law enforcement, by governments, and they can get access to that.

So Microsoft is one example, but there's also things like Meta.

There's also things like Google.

There's also companies that track where you go, location.

Maybe I've got a Fitbit and that tracks where I walk all day.

So there are some limits to this.

Courts have made some rulings about limitations of "You can't get ongoing location data "that's tracked via cell phone "because cell phones are considered at this point "to be essentially an extension of us "that you can't really operate without."

There are some limitations about the content of communications that can't be accessed that way.

There are some limitations about... Like, if I'm moving around on Google specifically, there is a lower court case that actively has said, "You can't reveal all of the location history "from Google because that counts as a search."

But those are all being challenged.

They're still active, and those are pretty minor limitations.

The vast majority of information, it has been ruled that you surrender that information, and so it does not count as a Fourth Amendment search for a government to request that.

And I say request explicitly because in many cases, law enforcement will go to these companies and just ask for the information without a warrant.

Sometimes there's a warrant involved, and that involves judicial sign-off and some evidence and some scope.

But in many cases, law enforcement will just approach a company and say, "Hey, we're looking at this person" or "We're looking at something in this area "from this time period.

"Can you hand us all of the information "that is relevant there?"

And some companies will, and they legally have the right to in many cases.

Brangham: Wow, remarkable.

Byron, there's another scope of this that we've seen more of recently, and that is sort of, I don't know, peer-to-peer or individuals in a way surveilling each other.

I think of Meta Glasses as the most recent example of this.

Again, for our audience who's not familiar with what those are, what are Meta Glasses and what are the issues around those?

Yeah, so Meta Glasses are basically eyeglasses that have little cameras built in and are either paired to your smartphone or internet connected in some other way.

And they've become controversial because it's a movement of cameras, not just the camera that we carry around in our phone, but it's being moved to our face.

And in many instances, people have been harassed by sort of online provocateurs who want to take videos.

Sometimes people don't know they're being recorded because the sort of warning light on them is very subtle or sometimes can even be disabled.

And so this is an extension of what we were talking about, right?

The ring cameras.

We've had phones in our pockets for the better part of 20 years now with camera capabilities and video cameras.

And some of that has been good, right?

Like, sometimes that has exposed real abuses.

Police officers who have been doing the wrong thing in the line of duty and have been brought to account.

Brangham: Or political movements in nations all over the world have documented abuses because they've got a phone that's internet enabled.

And it's been an incredible force for good.

Absolutely.

So there's lots of good reasons to have documentation and cameras with us.

But it's also changing a lot of the social norms.

And I don't think we as a society have fully caught up to the reality of everyone just having a camera all the time.

And when that camera moves to your face, I think that adds a whole other dimension to this issue.

Brangham: Are you wearing Meta Glasses right now?

Tau: I am not wearing Meta Glasses right now.

But I think in the next few years, we will increasingly start to see people with wearable devices of all kinds, whether they're glasses, whether they're little clip-on AI assistants or devices that record us passively all the time.

And those are going to raise a whole set of thorny, difficult issues that we as a society need to grapple with about what expectations of privacy do we have when we introduce something like a camera that can record, that can document, that can post something online in every moment of our lives.

I mean, Calli, would you just pick up on that?

What is...?

Just reflect on this idea, because we are in a brave new world, quite literally here, with this technology.

And, again, when I talk about this to people, I do sometimes hear a, "Come on, you're just... this is a little bit overblown."

The examples that you two have been describing, notwithstanding, people think, "I live a normal life.

"I don't have any secrets to hide.

"I don't care if I get seen shopping at Target "or that they know that I'm going to the pharmacy now "or that I typed "'best restaurant in Istanbul' into Google.

"If I live a normal, quote, unquote, 'legal life,' "what do I have to worry about any of this?"

How do you respond to that?

I hear that a lot.

And to steal a phrase from my colleague Jeremy Scott, I think one of the best ways to say that is that privacy constantly gets framed as having something to hide.

It's not about that.

It's about having something you want to protect.

So, when it comes to things like wearable devices, the risk factor does change, because when I hold up a camera, you can see I've got a whole camera in my hand that's pointing at you.

If I just have a wearable device in my glasses, if I have a pin that's recording, things like that, it's much more subtle, it's much easier to miss, and it's much more mobile.

So I can bring my Meta Glasses into your home and film a whole bunch of things you didn't expect to be exposed to the public.

I may be recording conversations of people next to me in a restaurant who did not think the person next to them was going to be recording and maybe uploading this kind of information.

There's also risks when it comes to recording children, recording things at playgrounds, in dressing rooms, in spaces that we've previously considered to be private.

Those are all real concerns.

And some of the concerns are not just with the devices themselves, although certainly glasses going into all of these places is a concern, but what that footage can then be used for.

So there's been facial recognition overlays that can find individuals from footage that's picked up on glasses.

There are nudify apps, unfortunately, that have been coming up all over the place online, where maybe you take an image of someone that you see on the street, think is attractive, and you're able to digitally create an image of them without clothing.

That's a huge violation.

That's a real problem.

Another issue we're running into is that when companies like Meta have been confronted with the risks of these devices, they're offloading legal knowledge and responsibility to users.

So for example, they're saying that you just have to be aware of what the wiretap laws are in their state, because some states, you're not allowed to record conversations unless everyone in the conversation consents.

You would have to, as a user, track whether your use of your glasses is legal in those places.

Brangham: I'm so sorry we have to stop here.

It's such an incredibly interesting conversation.

Calli Schroeder, Byron Tau, thank you both so much for being here.

Tau: Thanks for having us.

And that's it for this episode of "Horizons."

Thank you so much for watching.

We'll see you next week.

Narrator: Support for "Horizons" has been provided by Steve and Marilyn Kerman and the Gordon and Betty Moore Foundation.

Additional support is provided by Friends of the News Hour.

This program was made possible by contributions to your PBS station from viewers like you.

Thank you.

You're watching PBS.