Ideas - A rallying cry to extend human rights to our data-generating digital selves

Episode Date: March 11, 2025

In this digital age, we must think of ourselves as stakeholders, playing a vital role in the creation of data, says Wendy H. Wong. She is a political scientist and winner of the 2024 Balsillie Prize for Public Policy for her book, We, the Data. Wong argues for a human rights approach when it comes to how our data should be collected, and how it can be used.

It's not just you. Canadian politics really did get a little dramatic this year. And even people who don't normally follow or trying to figure out what the heck is going on. I'm Catherine Cullen, host of The House, and I started a new Election Weekly with two friends and fellow political nerds. I'm Daniel Thibault, bringing you the Quebec point de vue. I'm Jason Markazoff, bringing the takes and stakes from the West. Together, we are House Party, an elections weekly for everyone. We answer one big burning question every Wednesday. Find us in the House's feed wherever you get your podcasts. This is a CBC Podcast.

Welcome to Ideas. I'm Nala Ayad. Do you have any idea how much data you produce? You, dear listener, are a data generating machine, just by virtue of being connected to the modern world. just by virtue of being connected to the modern world. Almost all facets of human behavior have now become pieces of data that have become digitized. Mountains of data, galaxies of data, dense tangled webs of data, most of which we're not even aware of. So think about your sin, your address, your birth date. We kind of think of that as personal and very private

data. But that's just a tiny, tiny drop in the bucket of the types of information that are being collected about us. Wendy H. Wong argues that to be alive in 2025 is to be datafied. Whether it's gleaned by social media or fitness trackers, or our phones or facial recognition technology or whatever. Those data have a huge influence over our lives. And we don't have much say over just what data big companies scoop up, or how it's used to define or shape us and our behavior. Why I think data is a human rights issue is because there are not currently very many guardrails around how we use all these data that are about people.

Wendy Wong is the author of We the Data, Human Rights in the Digital Age. The book won the 2024 Ball Silly Prize for Public Policy. She came to write it after contemplating how the potential, promise, and peril of artificial intelligence rest on three things, computing power, algorithms, and data, the raw material that AI is built from. And a lot of the focus has just been on the algorithms or the compute. And I felt like the data part, that's the part that actually brings in the human being because the data are about what we do, what we think, what we're considering, where we're going and how we act.

And so the data part describes the human behavior that then the algorithms, you know, make predictions on the backs of the computing processes. So then when you talk about datification, can you explain what that is? It's a term that comes out of science and technology studies. But basically it describes how almost all facets of human behavior have now become pieces of data that have become digitized, so binary data. In other words, being able to be read by a computer. And that really changes how we might live our lives, both from understanding what people do, what people think, but also, you know, how our lives are being tracked,

so to speak, either in an explicit way, so when we use different types of apps, when we use our smartphones or other smart devices, there's a ton of data gathering happening, both in terms of what we put into those systems, like you post something on social media, but then the back end too. So, you know, our cell phones track what direction they're facing, where they are in the world, and other things like that that we may not be aware of, but that make our apps work well,

make the AI systems themselves work well. And so the datafication then is pointing to the fact that this is actually quite new in human history. We've always kept data about people. Ever since we started recordkeeping in a written know, clay tablets are a form of data. But it's never been so systematic. It has not been digitized in this same way. And it certainly hasn't been as widespread and pervasive in documenting really the minutiae

of human life. LESLIE KENDRICK-KLEIN The minutiae of human life. As you say, the collection of data about us is kind of integral to business, to government, operating. And throughout history, as you say, there have been technological shifts in our story as human beings. But I want to talk specifically about the way this shift affects who we are, our personhood. How is this era of

datification different from these other inflection points in history? I talk about this in the book as data stickiness. So there are really four ways we can think about why data are sticky. And when I say that, I really mean it like gum on the bottom of your shoe. And so that's the idea, which is that the data that are being created about human behaviors, the datification, is quite sticky. It's really hard to get rid of, and it's really easy to create. And what I mean by that is a lot of times the data that are being created are very mundane. They're not really extraordinary things.

So again, if you think about clay tablets or, you know, other written forms of data in the past, they would have recorded what we might think of as more important or significant things. But now, as I mentioned, the data about where you go in a day, I mean, think about our routines. They're not very extraordinary. They're mundane. That's why they're routines. The data are also sticky because once you make data, you're not keeping it in some little

data set in a lock box. You're actually sharing it across a whole bunch of other different types of data sets in order to derive insights. So they're linked and that's also what makes them sticky. The third way that they're sticky is that data are effectively forever. Which is to say that once we make digital data, how know, how do you know if it's ever deleted? How do you know where it goes? Again, because of its linkages to other data, its usefulness in the marketplace as something

bought and sold. So data actually are effectively immortal, which is very different from a human being. And the last way that data are sticky is because they're co-created. And this is something I talk about a lot in the book, because I think it's a real, excuse the pun, sticking point. We tend to think about data as belonging to someone, especially things that are perhaps personal or private in nature. But I make the point that digital data are only possible because there are a bunch of

people or companies out there, data collectors, who have an interest in learning something about human beings. And so they collect information from us, from all of us data sources. And without either party, you have no data. And I think that's really the point. It's not mine as a data source, and it's not necessarily the companies as the data collectors. It's a co-created thing.

So, I want to break down the answer you just gave me and talk about the components of that sort of bit by bit. And begin with how we are affected, how we are defined in fact, by this process of datafication. Are we kind of reduced to data? Is it that? Or are we multiplied into kind of an overwhelming amount of data or is it both? Yeah, I mean, I think that's a really good question. I think it's both. You know, like when people say I have a lot of data, I think it depersonalizes whatever

information is in those data. And so I think when the data describe human beings, human behaviors, human choices, we're flattening people out, right? People get taken out of the data, so to speak. We forget that the data are actually describing human choices and human behaviors because it's, quote, just data. And you know, data are really useful. That's why the collection of data has been so pervasive.

So what's the relationship between us and the data about us? Yeah, I mean, the relationship is that it comes directly from us. So rather than thinking about the data, about all of our mundane activities as data exhaust or as detritus, it's not a byproduct. It's actually descriptive of who we are. It's identity. And I think that that link has been lost. So in that sense, I think it flattens out human life.

I think it makes it easier to just say, this is what the data say, so therefore, you know, you suffer these consequences. Whether we think about that in terms of denying people mortgages or, you know, often it's used in assessing rental candidates for housing. You can think about it in terms of predictions for who might be a good driver, who might be a healthy person, in terms of the insurance industry. I mean, there's so many different ways to think about how the data then feedback into

how we are affected in all our different systems, social systems, economic systems, you know, to see certain things or be enabled in certain types of experiences, let's say. I wonder if it's too simplistic to say that we have become our data. I don't think it's too simplistic. I think there are a lot of people who actually make that point. I think it's important not to reduce human life to just data. I think we're both data and we are physical discretionary beings who make choices,

who have thoughts and things that happen in our lives that are still not to date tracked, but there's a whole lot going on in our lives that is tracked. So I would say it's a mix. I would say that we're increasingly being defined by data and that our life choices, what's available to us, is increasingly being squeezed by this extensive data analytic system we've created. And I sort of think that there's probably a distinction between what we should think of ourselves or whether our data is ourselves and how the big companies see us.

They see us as data, no? Yeah. I mean, they see our data and try to make decisions based on what they observe and what they pull from those data. Human beings are not discrete in the sense that we interact with the world. So how we think about things can change depending on what interactions we have. And so I do think part of what makes this identification challenging is that the more tailored or narrow those opportunities for us to interact are. I think the more it can reinforce filter bubbles, for example, on social media, what you see,

right? It can reinforce ideas in your head and make people more and more extreme and more embedded because they think that this is a widespread idea or this is something that everyone thinks instead of understanding that there's a whole range of ideas and opinions out there. And I think that's one consequence that a lot of people have observed. But I also think in general, the way that we live our lives through data and the data generation that all of us are engaged in with these data collecting companies, I don't think it's been looked at so much as a part of human existence going

forward. I think people sort of forget about that. And it's really easy to forget that 20 years ago, you know, datification was not so extensive. Right. It's kind of in the water that we swim in. Yes. And I think that's part of the sort of secret sauce, as it were, and what I'm trying to expose, which is that alternatives exist and that because the data are about human behaviors and affect human beings that we have a whole bunch of rights that exist, right? We have entitlements as human beings and this international framework around human rights

has been around since at least 1948. You mentioned the companies, of course, that are doing the data collection. What companies and entities are we talking about here? I mean, we like talking about big tech, the usual suspects, right? We love talking about like the Amazons and the Metas and Microsoft. But I think it goes beyond that. So if you think about the importance of data, the economies of data out there in the business world, I would argue that it's really any company that has a digital presence, any company

that is engaged online, which is so many. We interact in the world through multiple platforms a day. And so, really, it's a lot of companies are collecting data. A lot of companies are engaged in the processing and selling of data, and we just don't think about that. It's the usual suspects, of course. But it's actually a lot of data analytic companies you've probably never heard of. And we don't really interact with directly,

but they're more B2B companies as opposed to consumer facing. So when you say usual suspects, you mean Google, Meta, who else do you mean? So the big companies are Google, Meta, Amazon, Apple, Microsoft, and then sometimes like Tesla for example. So it really depends on who we're asking. You know, they're the Chinese ones as well, right? You've got Baidu and Tencent and so forth. So I do think that those are the ones we tend to focus on, but they're not the only ones

doing the data collection. They're not the only ones. But the ones you mentioned are some of the, like these are huge companies and really powerful companies and they seem to be, as we're seeing in the news even today, that they seem to be more and more self-governing and making the rules themselves these days. How are they different from the usual, you know, presence of powerful corporations that have always played outsized roles in our history? They are different in the sense that they are governors, which means that they create order around our lives.

And the reason they're able to do that is because of their control of different platforms that we're dependent on. So if you think about any sort of digital technology you engage in, there are only a certain number of providers. And so we don't have much choice. You either have a PC or you have a Mac, right? So that's either Microsoft or Apple. Think about the smartphone market.

You either have a Google-based phone, Android, or you have an Apple-based iPhone. So in that sense, you know, they're really shaping how we live our lives. So I think part of what's happening right now is that governments haven't come to terms with how exactly big tech has changed our lives because the focus hasn't been on the data that they gather and they are able to gather the data and analyze and improve their

products because of the platforms, because of these digital marketplaces where people come and interact and do things. And that's where all the data collection is happening. I want to talk about the purpose here of the data collection. A lot of people get very upset about this idea that these companies are using this data to either, you know, surveil us or to sell us something or to sell ads, you know, basically to sell us more things. But there's a bigger picture here, isn't there? What is that broader picture of what these companies or governments or agencies or whatever they

are, are doing with all this data that they collect every minute of every day? So yeah, I think there are some people who get upset about thinking about, you know, how their lives are being directed in sort of thinking about what we call recommender systems, so like how different websites will direct you to things or like you look at a pair of sneakers and those sneakers follow you across the internet in the form of ads. For years. Yeah, at least for several weeks, right? It's true.

And I think people have started to be upset about that. I think some people, though, don't care about that either. Okay, and that's fine. That's sort of like a preference thing. I do think, though, it's worth thinking about the fact that there are so many data being collected about us. And so the potential for the use of those data, because they are effectively immortal, is infinite. We just don't know. And I think part of what my concern is, why I think data is a human rights issue, is because there are not currently

very many guardrails around how we use all these data that are about people. So we tend to think there are certain types of data that are, quote, private, right? So think about your sin, your address, your birth date. We kind of think of that as personal and very private data. But that's just a tiny, tiny drop in the bucket of the types of information that are being collected about us. And so if someone knows, you know, your daily routine, let's say how often you take public transit or that you walk to work or that it seems that you're biking to work and that

you go to the gym every day at 5 o'clock, there are so many inferences they can draw about you that I think constrain not necessarily privacy, but your ability to choose. You know, that means that if people can predict where you're going to be or how you're going to act in a certain way, I mean, if there are problems in the future that you haven't anticipated, that could open up the door for lots of different ways that our choices can be constrained going forward. That sounds really abstract, but I do think that one of the things that folks tend not to worry about is stuff that doesn't seem important at the time.

You know, that apocryphal Target story of the girl who, you know, her purchases at Target reflected that she was pregnant before she told her parents. You know, her parents find out because Target ends up showing her or giving them advertisements and coupons for a baby, right? That's how they find out. I mean, in a way, you're like, okay, well, what is the big deal between my purchase history and what I do in the future? Well, it's that there are things that we might want to have control over, like revealing whether or not we're pregnant.

And the machine doesn't necessarily have the discretion we do. And I think that's really the concern. So I understand, though, when you talk about this bigger picture of what these companies are doing with our data, they're also using the data that they gather from us, as you say, our preferences or what we're planning for, to create new data about us. Mm-hmm. Or new tools, right?

So, you know, this entire explosion of chat GPT and other large language models is only possible because of data that these companies are effectively vacuuming up from the internet. So if you recall, at the time ChatGBT dropped a couple years ago, it was, oh yeah, we just took all the information from the internet because it was just there. And again, this is another one of those unanticipated consequences. Did bloggers think that, you know, you

blog about something, did they think that 10 years later some bot was just going to go and suck up all that information to sound more like a human being in the way that they interact with people? I mean, that's sort of, it was there, and so it was taken. The example of chat GPT is really important because on the one hand, these large language models are incredible. They sound decently human. They can help us do things more efficiently and more, perhaps more effectively. On the other hand, they really hide the fact that they're built on tons and tons of data that people have worked hard on, that they've put their lives into.

And it becomes, you know, an essay that might take six months for someone to write. It's just sucked up into, it's one of trillions of examples, right? So how do we think about the human in those data? I think it's really important to point out that, hey, all these data that are in these systems that are creating these advanced computing products. They started from somewhere and that's from all of us. And we need to really think about what that means for us going forward. Wendy Wong is a professor of political science and principles research chair at the University

of British Columbia, Okanagan in Kelowna. And she's the author of We the Data, Human Rights in the Digital Age, the 2024 winner of the Ballsilli Prize for Public Policy. Ideas is a podcast and a broadcast, heard on CBC Radio 1 in Canada, on US Public Radio, across North America, on SiriusXM, in Australia on ABC Radio National, on World Radio Paris, and around the world at cbc.ca.ideas. Find us on the CBC News app and wherever you get your podcasts. I'm Nala Ayed. and wherever you get your podcasts. I'm Nala Ayed. Canada is facing a near unprecedented set of political challenges right now. I'm Jamie Poisson and on Frontburner, the daily news podcast that I host, we're covering those stories five days a week. We try and bring clarity to Donald Trump's overt hostility to

our country and how that has thrown a wrench into our looming federal election. So if you're looking for a show to help guide you through an increasingly chaotic world, find and follow Frontburner wherever you get your podcasts. Which data is more personal and sensitive? Your social insurance number or your heart rate? The photo on your driver's license or your kid's first day of school pictures that you posted to social media. And that's just the tip of a massive iceberg of datafication,

most of which we surrender willingly or unwittingly whenever we get a new app or gadget or search for something online. And how much do any of us know about how that data is used? It's mostly out of sight and out of mind because all this data collection happens in the back end. That is, the machinery that makes apps, websites, and platforms work. It's also mostly out of our control.

And Wendy Wong wants to change that. You know, the Europeans have made some moves. You know, the general data regulation, protection regulation has done some work in terms of thinking about what rights one might have over the data that describe them. The problem with the way we think about data, however, and sort of our entitlements as individuals is that you have to be identifiable in the data. And most data that are collected about people are not identifiable in that they don't attach a name to those data, right?

And so one problem is you can't even locate yourself in a data set. The other problem is simply that, you know, if it's not identifiable, they really can't pull it out, right? So in that sense, we have this real analog sense of data propriety. You know, if I can't figure out that it was, quote unquote, mine, then I can't take it out. I think datification has created a situation where it almost doesn't matter that you can't attach a name to those data because even if I'm anonymized in one data

set, you know, the fact is there's so much data tracked about me as an individual across data sets that it's actually fairly straightforward to sort of figure out and re-identify people in anonymized data sets. And there are lots of different examples of how this has happened in Australia when the Melbourne Transit System had a hackathon about users of their transit system and the data were anonymized. But, you know, a bunch of data scientists got into it and figured out pretty quickly, you know, a whole bunch of the people who were in that data set. Wow.

They were able to identify individuals. Yeah, just by pulling pieces together. Like knowing, you know, maybe they start by identifying themselves. Like I took the transit system on these days at these times. And then they can match that up with other people they might have known who rode transit at the same time. And there was an elected official who tweeted about when he had taken transit. So they just took the information from the tweet and, you know, you can triangulate. So we're not safe in anonymized data.

I would say that to be safer, to be more respectful of human rights and our autonomy, our dignity as people, we need to be very careful about the types of data that are even allowed to be collected. I think that's where we start from. If the data don't exist, then it can't be fed into an AI system or any other computing system. So I do think, to answer your question, as human beings, as social groups, like we should have more say over the types of data that are being collected

as co-creators, but we're not being asked, nor do I think governments are really thinking about data. I think they think about AI systems. Right. You referenced the European example, and they're sort of further ahead on some of the stuff than the rest of the world. And this whole idea of signing on to a company's terms and conditions, you know, suggests some kind of agency. I mean, that's debatable, of course. But how much input

do we actually have if we have to agree to the company's terms and conditions if we want to use the technology? Well, I think we don't, right? I mean, I think we are excluded. You know, I've been talking about this book quite a bit for the past couple of years. And you know, I'm always surprised by the number of people who come up afterwards and say, you know, I'm so glad I didn't choose social media or I don't post anything on any of these social media platforms and that's sort of how I've stayed away or out of the system. And even if you're not on social media or you don't have certain apps, great.

But at the same time, you have a smartphone. Cars are notorious tracking devices as well, right? So they're tracking the way we drive and what you do while you're in the car, in part for safety reasons, in part for other data gathering reasons. But like the way that we can't opt out, you know, it's not just what we choose to do, it's also what other people choose to use. Are companies using facial recognition to do marketing research? Are they, you know, tracking your behavior online because you went to their website and you allowed for cookies?

We cannot opt out. And that's because datafied systems are so pervasive. And I don't want to say they're all bad. They're not bad. They help things run better. On the other hand, I think the spread of these systems has really become such that it's hard to get away from them. And it's certainly hard once you think about their scope, how much data they have about

all of us. So, but you know, not only we can't opt out, but is it not true to say that we can't fully participate in our society today if we're not being datafied? If that's the right verb. I think that's a really good question. I'm not sure that you could. And I'm not sure that if you, you know, live in conjunction with other people, if you are, you know, going to church or going to work or going to school or engaged in commerce, I don't think you can get away from these systems. I really don't because other people have chosen to, you know, manage their customer databases in a certain way.

And so these are back-end functions that we may not even see. And it's only when there are mistakes that we can see that. So there was a case a year or two ago about Home Depot and its relationship with Meta and how, you know, they were supposed to not be sharing customer information with Meta, but they were. And like somebody found out through a blip. And so it's hard to let go of the skepticism, I would say. Like, so when companies say, oh, we delete your data, you know, this is very common, people say, okay, well, it's okay, because they agreed to delete the data. And my response has always been, yeah, but the data are effectively immortal and they're shared.

And so maybe the company has deleted that data set, but have they guaranteed that the third parties that they work with have deleted that? I don't know, right? And so again, my answer is create as little data as necessary for the functioning of our societies, our systems, our economy, and try to think about what the human cost of making all these data really is, which I think is really hard to say outside of very specific examples. But knowing that these examples exist means that,

even the most mundane things like checking out at Home Depot, the data about your transaction may be linked to your social media. So you also talk about a solution in which, you say that we should become treated as stakeholders in the data that we co-create, and also in addition how it's actually used. Can you talk about in practice what that would mean? What kind of shift would that require? So, I actually think we should think about ourselves as stakeholders.

So part of that is if we understand that we do actually play a really important role in the creation of data, that we are half the equation, I think that gives us, I mean, it sounds really airy-fairy to say just shift your mindset, but I do think that actually really matters because if you think that you have a stake, if you think that you're just as important as the Googles and the Meadows of the world in the datafication of our lives, then that gives us agency in democracies to make demands of our leaders and to tell our policymakers that these practices are unacceptable. But are we really half of that equation?

Can we equate ourselves with Google and Meta and Twitter? Sorry, X. No, no. So yes, I mean, by definition, to collect data about an individual person, that person must exist, right? In the world where the data are about living people or actual people who have existed. Now, you're absolutely right to point out

that any one of us is no match for these giant trillion-dollar market-valued companies. But I would say that if it's just me or you having this realization that we're stakeholders but nobody else thinks that way, well, yes, of course. You know, this is why as a political scientist it actually really helps to think about these issues because we think about collective action and what it takes to change a situation. And it's not going to happen because one or two people believe in it. It might happen if a collective believes in it. And then what are they rallying around? How are they going to think about themselves as part of the process, as stakeholders? I think that's where you see all social movements, all changes, the demands for change that come from citizens, that come from the grassroots, they all come from people realizing that they

are different from what they've been told they are. So if you say you're a data stakeholder, it means that you have a reason to even be part of the conversation. And I think that's really important. I think that's the start of any movement, whether we're talking about labor or women's rights or LGBTQ rights or indigenous rights. People have to realize they're part of the process and that therefore they can demand changes.

And this is also why I talk about data literacy in the book, is that many of us are intimidated by this idea of data. And so if data could be demystified and understandable for your common everyday person, that is not a scary thing. You don't have to be a data scientist to understand that data by their existence are biased because someone has to make data. Data don't exist in nature. I can't because someone has to make data. Data don't exist in nature. I can't go outside and grab a data. I can go outside and grab a leaf, but not a data.

But to focus on the rights here, our rights in the context of big data. Often when we talk about, even as we're talking right now, I think in my own case that I worry about privacy and being free from surveillance when you're online or you're on social media, and also not just allowing big tech companies to vacuum up all our data and use us as the product. But can you talk about what are the human rights that you want to bring to the discussion of datification? So, yes, I think a lot of times when the conversation around human rights and AI comes up, a lot of people say, well, privacy, privacy is really important. Or sometimes we talk about freedom of expression, right? And I think those are the two that are very surfaced in this conversation because we can

feel their effects. But I would say actually, we need to step back from these very intensely discussed rights because actually, datification, if it fundamentally changes how we live, it affects not just your privacy, it affects the way you access education, so freedom of education,

it affects our knowledge and access to healthcare, so right to health, it affects your freedom of education, it affects our knowledge and access to healthcare, so right to health. It affects your freedom of conscience because if you're sort of guided down a certain algorithmic path and not offered any alternatives, how does that affect the way you can freely think or resist or question? Pretty soon, if you go down any sort of list of human rights, you can think about how datafication, how automation and AI systems might change how we enjoy those rights. So I choose to think about it in terms of the values of human rights. And it really comes down to four.

It comes down to ideas around autonomy, so around choice. Comes down to questions of dignity, just how we are treating each other and how we want to be treated as human beings. It comes down to equality, which is this idea that we're all compared to the same standard, you know, people aren't held to different standards. And finally, this idea of community, thinking about who we are as individuals existing with others in society. What does that community look like?

Is that community polarized? Is that community, you know, a majority that might inflict harms on a minority? What does community mean to live with one another? And I think those are the four values of human rights that are actually actively being challenged. You know, I talk about several issue areas in the book, and one of the issues is facial recognition technology. So we can just think about how systems that take data about our faces to verify our identity, to try to identify one person out of many people, like in terms of like, in crime context, for example. These technologies, they challenge

our autonomy in the sense that we often can't choose whether or not we're in those systems. There are companies like Clearview AI that have scraped the internet for all kinds of faces. They have billions of faces, billions of examples in their data set, for example. None of us have opted into that, so that's an autonomy issue. It's a dignity issue because we're talking about faces. These are the things that actually make us individuals in the world. And so the data about them are being taken from us and being used in ways

that sort of treat us and our faces as objects as opposed to people. We can think about equality in that context because we know facial recognition systems are disproportionately used against minoritized populations, against black populations and indigenous populations, against refugees, against immigrants, right? These are populations that are disempowered already. And we can think about community. What does it mean to have this kind of capability where people are able to be surveyed in this way, to be tracked, to use something as personal, as mundane. It's who we are and it's being used against us and that really speaks to

how we live as a community. Staying with that same example, you know, if we held human rights kind of in our minds as a framework for thinking about, you know, face recognition technology. How would that change how data are collected or used or how we participate in this data economy? I think we would be much more choosy about how we use facial recognition. I mean, you know, Air Canada is using facial recognition to help with plane boarding, right? And one of the big examples when I was writing the book was Cadillac Fairview got in trouble.

Cadillac Fairview owns a bunch of shopping malls throughout Canada. They were using facial recognition for marketing purposes to figure out the demographics of people who are going to the mall. I mean, do we need that technology for those purposes? You know, I want to hear the public discussion. I want to hear, you know, how companies are actually made to justify their use of this technology. Like, in this country, we have protections around biometric data and facial recognition

or facial data rather is part of that. You know, it seems that it's pretty tough for the government to really try to enforce these protections around biometric data, right? Like, the facial recognition systems are being deployed before we have a set of norms around what it means to use someone's face as an identity measure, as a crime prevention measure, as a security measure, under what conditions should we be allowed to use facial data? That conversation hasn't happened. What is that, Wendy? This is a question I think about a lot, and I'm wondering what you think. What does that say about our moment that such ground shifting technology is deployed

before we've had those conversations, not only deployed, but almost entrenched in our day-to-day lives. I think it says we are at a moment of extreme change that is not that different in the sense that, you know, it's hard for governments to make rules when the technologies don't exist. I think we have a serious misunderstanding of the technology that we have in hand. So AI as a class of technologies I think has a lot of potential upside. On the other hand, the data that go into these systems is rarely the object of conversation or debate. The

data are almost a foregone conclusion and I think that that's a mistake. Because if we don't understand the importance of that data for making these systems useful, if we don't understand that those data come from human beings who are alive, I think we have a real problem. We misunderstand the potential harm. And I do think that digital technologies that rely on data, that AI systems are part of these technologies, they pose a different kind of challenge than previous big sectors. So we think about big oil or big auto or big pharma. These are big sectors and they are rich sectors and they provide a lot of value for human beings. They're not as pervasively spread.

They aren't in all facets of our lives. Some of them do, you know, undergird a lot of human functions, but they're also not so minute in their involvement in our lives, right? They're not about the most mundane things. And we can also, in some ways, opt out in an active way that we can't actually do very well in a digital data framework. So it's just a far more overwhelming policy question than we've ever seen.

Yeah, and this is why when I, so when I first started working on this book, it was right before the pandemic, a lot of the writing at the time was from the people who were making the technologies or it was from philosophers who were thinking about, okay, well, how does this fundamentally change how we live? Or some economists were thinking about the market changes. But if we are to think about a set of technologies that has such fundamental shifts in both perhaps the way we live, both from a positive and maybe an existential negative way, how can we try to regulate these kinds of technologies without fundamentally understanding what enables

them, which is again these three components of AI, data, compute, and algorithms. And I think we focus so much on compute and the algorithms without thinking that data is another lever that governments can pull on to actually pull back the effects of some of these companies. And in fact, governments are quite good at data. They, you know, because they collect a lot of data about people. They used to be, you know, the ones who knew perhaps the most about any given population within their borders and, you know, big tech has overtaken them. But I do think that governments spend a lot of time thinking about what kinds

of data are too personal to share widely versus data that should be made available because it has some sort of social good or social function. So we've talked about the power of big tech, where data is concerned, and I'm wondering what you think the prospects for human rights in the realm of big data, given that we've had this US election and the current climate that has come with it, and the thinking about deregulation that now exists, what are the prospects of us actually incorporating human rights as a framework of looking at this process of datification? Uh... I'm going to try to gather my thoughts here. Yeah, take a moment.

You know, I think we're at a moment of flux, and, you know, at the time of this conversation, the Trump administration has just started, and we are faced with what feels like a barrage of changes on a daily basis. At this very moment, it's hard to answer that question with any sort of positive framing, I think. I don't actually think the moment has changed that much in the sense that what we were talking

about earlier, governments have been focused on developing AI. I mean, you know, Prime Minister Trudeau said he's in support of nuclear power plants to channel resources towards the development of AI. You know, that's part of the compute equation. Right. Well, we're focused on AI safety in the sense that we need to make algorithms transparent and understandable, well that's fine, but that's not a question about data.

So in a way, you know, the political mouse from right now around all these different changes that the United States is trying to implement is obscuring the fact that I think that governments have not been thinking about data when it comes to AI in particular, and that has not changed with the change of administration down south. You, in your book, make the point that big tech is really infrastructure. Mm-hmm. I think just going back to, you know, thinking about how they make things work.

So we don't think about how electricity enables our conversation right now, our ability to stay awake past dark, to do work, to eat food, to interact with each other, because it's embedded in our everyday lives. It's become the infrastructure of human life. And in that sense, I think datification and increasingly now AI systems have become embedded in our lives in the sense that we don't even know how they are acting or we've gotten used to how they act on us. And that's how we've now shaped our lives, right?

That's how big tech companies govern how we live by creating the order that constitutes what it means to live. So what would that mean for, again, for thinking about human rights and the responsibilities of big tech, knowing, as you say, that they are part of the water in which we swim? Well, it means that we have to take quick and serious steps about thinking about the responsibilities of companies in terms of human rights. So traditionally, human rights have not applied to companies directly. States are responsible for protecting human rights.

So if a company violates human rights within a state's borders, it's actually that state's fault. I think we really need to rethink that. One, because the potentiality for harming human rights, especially in terms of the values I talked about, autonomy, dignity, equality, and community are so great, especially when we think about tech companies and AI systems. But also because I think it's a lot of responsibility for governments to take this on in the sense that I think there's just too much work if we direct everything at governments to manage

the companies on their own. I really do think that when it comes to human rights and harms against human beings that we should be directing more scrutiny and more responsibility to big tech companies. If they want to be more active in our lives and really shape people in their lives the way they do now, they should also have the responsibility, which is what governments have, especially democratic governments, have dealt with. How do you think the prospects for human rights in the realm of big data is even possible

at a time where writ large, you know, human rights are, you know, not to be simplistic, but in some senses are out of fashion, you know, both in the national and international context. I mean, does that, how does that play into all of this? It's never a good time for human rights. Since we created the system of human rights protections, there have been violations, right? And just because we made law around protecting human rights

doesn't mean that torture doesn't happen or torture didn't happen. And I think there is some cynicism around that. But I also think, and I try to remind my students of this all the time, which is that just because there's a rule doesn't mean there's complete compliance. And so that's why we need the rule to show us what's right and what's wrong, what falls above the line and what falls below the line. And if you don't include data in the sort of scope of human rights, then sure, I mean,

you could say maybe you think it's not effective to do that, but then you've also not created that sort of bright line that says, hey, certain types of uses of data are fundamentally problematic from a human rights perspective and others are not. And I think if you don't set the rule, then you don't have any basis by which to critique. And I do think right now we're struggling with that language. I think that's why people have focused on privacy or focused on freedom of expression because those are tractable.

They're not abstract. But I do think a lot of the reasons why this conversation around data and human rights feels abstract is because the data collection and use is happening on the back end. And so people don't experience it in the same way they might be afraid of or experience political repression. You argued that for people to become empowered to take back control of their data and to prioritize human rights, we need to become empowered to take back control of their data and to prioritize human rights, we need to become data literate.

Can you explain what data literacy entails exactly? Sure. And I actually said we need a right to data literacy, which the more I thought about it, the more I thought if we don't elevate data literacy to an entitlement, it becomes an option when I think it's not an option. And so the reason why I think data literacy is so important is because datafication is not going anywhere. I don't think we're going to stop making data about people.

We might restrict the types, but I don't think that datafication's actually going anywhere because there are good uses for it. And when I say data literacy, what I'm really trying to point to is, if we're all living in a situation of datification, we have to understand how data work, not as experts. So I don't think everyone should go into data science, far from it, or I don't think we should all be working with data on a daily basis. But what I mean by that is we have to understand collectively how data are fundamentally a

human artifact. And because they're human artifacts, what are the biases, what are the values, the preferences, the choices that are embedded in what seems like a neutral data set, right? What seems like a bunch of numbers on a screen is actually the product of a lot of different human choices. What you think is important to observe, how you choose to record that information, and how you choose to analyze and use that, what becomes data, is a series of human decisions. And so becoming data literate means actually understanding how to make data.

I use the example in the book of choosing to collect data about leaves. So you can make any kind of data set around leaves. Are you going to talk about how big they are, what trees they come from, what color they are, what their texture is? There's a lot of different choices that anyone could make in the process of making data. And I think to do that, to have that kind of knowledge and framework around understanding this is why I'm seeing the things I am because someone else has made these choices about what they think is important to me or to their business or to society, I think that's a really

important critical step because we do often take the information we get for granted because we build in ourselves some assumptions about the veracity of that information. And I think we need to come to terms with the fact that data have their own biases. But does it mean we have to read all these, you know, 23 pages of terms and conditions before we agree to them to use an app? Like, what's the process of apprehending the scope of datification and our role in it? I think we should teach people to make data. You know, one of the reasons why I think maybe I'm comfortable with data as a social scientist is because that's part of my job is to work

not just with data, but with human subjects' data. And so as a social scientist, you know, every time I want to talk to people or gather information from a living person, I have to go through a series of questions that ensure that I understand the value of those data and to ensure that I actually need that kind of information for my research. And I think the process of making data about people has made me attuned to the humanity attached to those data. So I think part of it is just most of us don't consciously make data sets. So if we're taught in K through 12 education to think about how data are made, to actually

make data sets, whether that's digital or just sorting things and articulating how we came to those conclusions and how we came to those data. And I think in the book I talk a lot about the use of libraries as conduits for helping people understand data and to become data literate because that's what libraries have always done. Libraries are fundamentally going out in the world, sourcing materials of public value, and then organizing those sources for people to access.

That is the point of a library, to make things coherent and to store knowledge in a way that users can use. And I think that that is fundamentally an analogy to thinking about data. How are big tech companies making data? Why do they make the data? They want to make order out of the noise of life. They want to get analytics to understand, you know, all the things that are happening in the world so that they can make sense of it and either make products or convince us

to do certain things or what have you. So I think that data literacy as a set of understandings, and this is why it's literacy and not just knowledge, it's being able to understand the world from a data perspective that is really important. And I don't know if listening to me is convincing enough or in depth enough for some folks to really get that. It certainly is for me. And Wendy, thank you so much for taking us, I guess, to the start of the journey to data literacy. I really appreciate it. Thank you, Nala.

Wendy Wong is a professor of political science at the University of British Columbia, Okanagan, in Kelowna. Her latest book is We the Data, Human Rights in the Digital Age. It won the 2024 Balsillie Prize for Public Policy. Special thanks to Alia Ramadhan at CBC Kelowna, and to the Writers' Trust of Canada, which presents the Balsilli Prize. This episode was produced by Chris Wadskow. Our technical producer is Danielle Duval.

Our web producer is Lisa Ayuso. Senior producer, Nikola Lukcic. Greg Kelly is the executive producer of Ideas, and I'm Nala Ayed.