Toomas Hendrik Ilves, former president of Estonia and 2018-19 CASBS fellow, discusses recent technological attacks on democracies around the world, and how we can address them.
Movement for obfuscating the Electoral College in the US
First-past-the-post electoral systems
and a glossary of alternative electoral systems
Atlantic Council’s Digital Forensic Research Lab
Open source investigative platform Bellingcat
Transatlantic Commission on Election Integrity
Alliance for Securing Democracy, Hamilton 68
“The Two Cultures” C. P. Snow
“Estonia, The Digital Republic” The New Yorker
Stuart Russell on algorithms from “John Brockman: Possible Minds” The Long Now Foundation
Center for Advanced Study in the Behavioral Sciences
@casbsstanford on twitter
Narrator: What's your goal here?
Toomas Hendrik Ilves: Preserve democracy for the digital era.
John Markoff: Welcome to Human Centered, a series of short conversations with researchers at the Center for Advanced Study in the Behavioral Sciences at Stanford University. The center was founded in 1954 to encourage interdisciplinary research focused on the most pressing societal issues of our time. Each year, a range of scholars, scientists, and government officials come to spend a year studying contemporary societal problems. My name is John Markoff. I'm a science and technology writer and former reporter for The New York Times. In these conversations, we've set out to find interesting projects at the Center that shed new light on the way we think about society. Today I spoke with Thomas Hendrik Ilves, the former president of Estonia, who this year is a Berggruen Fellow at the Center. He has been focused on regulatory changes and technologies necessary to ensure the survival of liberal democracies. Ilves is a frequent advisor to governments around the globe who are trying to emulate some of the advances that Estonia has made in digital governance. We spoke about the challenges that internet technologies have created and their impact on the United States and democracies around the world. What is the project you're working on?
Toomas Hendrik Ilves: Well, again, I'll talk about how I got there, which was that I was already worried about the use of the internet or of digital technology to manipulate elections for about a decade. But in 2016, it became very clear, and I was one of the first people to actually write anything about it. Nothing big, just like saying, "Hi guys, this is going on." Then I've spent the time since I got to Stanford in 2017 basically researching everything that has gone on and what can be done to mitigate these problems. And sort of sitting up here looking across the bay, it struck me one day that in fact you can tinker and you can come up with maybe new algorithms to fight deepfakes, or you can do something to identify fake news, or you can find something to identify bots, but It will always be Achilles running after the tortoise. You will never catch up. And that really the fundamental problems we face are the vulnerabilities in democracy, the way we've practiced it since basically the 18th century, that there are certain weaknesses that need to be fixed. Now, the most sort of glaring one, I'd say, which is here in the United States, is the Electoral College, where you can actually focus on certain key states because of the disproportionality or disproportional weight, and this way have a maximum effect. But more broadly, it's not just that, which is strictly a US thing. I think that first-past-the-post electoral systems, or also known as single-mandate electoral district, a congressional district, and whoever wins, wins. That leads to several things which increase the possibilities for manipulation and also for polarization. I mean, first of all, a first-past-the-post electoral system, as you have, say, in the US and the UK, leads to a two-party system. The two-party system leads to a middle ground that is more away from the center than you would expect. As opposed to what you have in most countries in Europe, which is a multiple mandate electoral district. And there you have— I mean, you say you have a district, electoral state, for example, and you have— you may have 100 people running, but 5 of them win. But generally what you'll end up having is kind of the centrist will win, and you also have some outliers either on the far left or the far right, maybe. but no single party in that kind of system ever wins, or very rarely do you have a situation where one party wins, gets over 50%, which then leads to a coalition. This leads to politics in Europe that is more centrist-oriented, less divided. You can look at— I mean, where are the divisions the greatest right now? At the US and at the UK right now, where you have this complete sort of breakdown where you have, in the case of Brexit and the Remain and the Leave people, I mean, they're not talking to each other at all and you just have this huge chasm. Now, that's another example of where you're more susceptible to propaganda than you are if you have multiple-mandate electoral jurisdictions, but in the case of Brexit, it was a referendum that did it. And again, referenda, where you have binary decisions, you are very susceptible to outside interference. And so we are only now finding out, and with sort of tremendous reluctance on the part of the authorities, that of the amount of manipulation that went into the Brexit referendum with not only the lies about its effects spread by people running or who were leading it officially, but the amount of sub rosa manipulation that went on on the part of Russians buying ads or paying or manipulating people to buy ads. I mean, there are people who were buying ads probably with not being transparent about it, but at least they were UK citizens, but apparently they were offered deals for like diamond mines. We saw actually where it really sort of came full force was in 2014 with the Ukrainian— well, even before the actual toppling of Yanukovych, I mean, you saw this narrative being pushed about, you know, the Ukrainian fascists. Then you had fake videos that were supposedly news that were— these were taken up by a rather naive Western journalist. So you had these stories about, you know, sort of Ukrainians supposedly sort of crucifying a Russian child with this I mean, it turned out it was an actress. And especially the BBC was very much into, well, on the one hand, the Russians say this, on the other hand, the Ukrainians deny it. And this naïveté with regard to Russian media reportage did a huge amount of damage for the Ukrainians back then. I mean, now, 5 years later, even a couple of years afterwards, news sources have become more sophisticated. I mean, that's also different from what is posted on Facebook. That remains the same, but at least the Western news outlets have become a little more skeptical with regard to these things and don't report them as if it were all true. So that's why I've been dealing with it for a while. And of course, the full panoply of possibilities became apparent in 2016, everything from hacking to doxxing, the publishing the materials that had been hacked, to the manipulation of news, fake news. I mean, a word I don't like, but basically using pseudo-journalistic outlets, maybe from Macedonia.
John Markoff: Did you get an inkling of any of this in Estonia? You said you had written about this early on. Was it from things that were happening in the digital world in Estonia that alerted you to this?
Toomas Hendrik Ilves: Well, we had the massive DDoS attack in 2007, which I say usually that, well, this is the first sort of case of outright cyberwar in a sense, because it was the continuation of policy by other means. In order to pay back Estonia for moving the statue, they did a massive DDoS attack against us. And that was actually the first time I said, well, I mean, you can in fact use this for political ends. And then—
Narrator: Do you see any of the steps that Estonia has taken to have inured against some of these things?
Toomas Hendrik Ilves: Well, some do. First of all, in my country, since the beginning, you cannot transfer property with an anonymous— you cannot be an anonymous shell company. And that's been there since basically 1991. But it is also one of the 3 transactions you cannot perform digitally. You have to show up for that. So getting married, you have to show up. You have to show up if you get divorced. And the third is you have to have a member of an identifiable real company show up if you're going to buy or sell property. And we did it for completely different reasons back in the early '90s because we said, okay, we're poor, we're little, and like, you know, you could just buy up, you know, neighbor to the east is going to buy everything up. But now looking at, say, the UK, US, Trump Tower, I mean, these— all these properties you can buy in all these countries. I mean, the UK is the same, huge problem. Then it turns out that the children of various authoritarians in not only Russia but other authoritarians in the former Soviet Union, they're all living in London. So that's one thing. Clearly, we have— I mean, our system is ultimately based on having a strong digital identity, and that gives you a lot of security. It doesn't work against the fake news stuff, but in terms of making an efficient governance system, you really do need a strong digital identity with, you know, end-to-end encryption, two-factor authentication. You can sign documents digitally, legally, and it has legal force.
John Markoff: To this issue of democracy and technology, do you find yourself able to remain optimistic?
Toomas Hendrik Ilves: Cautiously. I'm optimistic in the sense that we will find a way to deal with these issues because there is nothing short of a round-the-world electromagnetic pulse that is going to stop this. And there will be enough people with Faraday cages around their computers. We will have to deal with this. That's why I've come to the conclusion that we need to, in one way or another, address the vulnerabilities of democracy or democratic procedures that were from the 18th century. I mean, the Electoral College was designed that way, if you read the Federalist Papers, basically to make sure that 3 big states—Virginia, New York, and Pennsylvania— didn't dominate the process. Whereas that's, you know, we're living in an era in which, you know, presidential candidates just fly out, fly back. I mean, everyone gets to see the candidates and it's all— and we live in instant communications.
John Markoff: You know, I've seen there's this movement, the winner-take-all movement, that would sort of minimize the Electoral College impact. Do you think that's How fixable do you think the Electoral College is?
Toomas Hendrik Ilves: Ultimately, I think that there's enough gut reaction against something that has now twice in the past two decades gone against the popular vote. It just seems wrong. It'll be more difficult to deal with things like first-past-the-post elections, but I think that if we end up having enough bad results results that people start thinking about them. Referenda, I don't know what to do. I think for referenda you're going to need to, well, first of all, limit on what you do referenda. And then maybe offer more than one choice, not make them binary choices. Because it's, I mean, you were just saying, you know, here I am, I'm vulnerable, you know, use whatever you want. That is something we can't allow. Of course, there's a hugely asymmetric relationship with the people doing this because they're not democracies. I mean, if you're dealing with an undemocratic country, you're not going to affect them this way. You have to think of other ways. I mean, you have to basically harden your defenses. But I do keep thinking that basically we are— digital era that we're in is as fundamental a change as, say, the invention of the printing press.
John Markoff: And yet, so this era is something that's actually fundamentally new.
Toomas Hendrik Ilves: Is that sort of your sense? I would actually claim the digital era began in 2006, 2007 with social media in the form of Facebook mainly then and the smartphone coming out and the decision to take social media from the PC to the phone, which led to like this incredible hockey stick of a graph and just like took off. And the sort of ubiquity of the smartphone and the ability to monitor things which previously were not puts us in a new era. So I'd say, I'd say I'd say it's 12 years old. Do you have a name for it? Well, I call it the digital era, I mean, as opposed to the Gutenberg era or the automobile era.
John Markoff: It didn't really happen until it became personal, is your point. Right.
Toomas Hendrik Ilves: I mean, we had computers and I was, you know, I'm one of these, I learned to code 50 years ago. I mean, BASIC, I admit, BASIC is very basic, but it didn't become this ubiquitous worldwide phenomenon until the smartphone came. With all of the data that can be accessed.
John Markoff: Stuart Russell, who's an AI scientist, articulated this mechanism for driving people in different extremes that was not very sophisticated. It was chilling in its simplicity. It was an artifact, he argued, of the system that the social media companies have created to create engagement, which causes a feedback loop which moves people in a direction toward an extreme. And he said, you know, it's 50 lines of code and it wasn't the intent, but it's basically, it rewires people politically and moves them incrementally in the direction of—
Toomas Hendrik Ilves: I know him. He's great. And I have his— down in my other office I have his book. He's been also worried about— Weapons. Weapons.
John Markoff: He's big on weapons, yeah.
Toomas Hendrik Ilves: It's one of those mechanisms that we have to develop resilience against via our democratic electoral procedures. So, you know, I was—
John Markoff: this digital identity question, I was just absolutely startled to see the extent of the Facebook challenge with fraudulent identity. They acknowledge that they, in the last quarter, took out 3/4 of a billion fake accounts. Yeah. That means to me that they don't have their hands on the problem. If there are that many accounts being created, there's got to be an economic— something's at work that's making a lot of money or something else.
Toomas Hendrik Ilves: Well, 3/4 billion, that's like 20% of their— Yeah, yeah, it's staggering.
John Markoff: That's what they acknowledge.
Toomas Hendrik Ilves: Well, I mean, since it's all borderless, it's tough.
John Markoff: Have you had the opportunity to engage people from Facebook? I'm looking particularly at the Facebook quadrant of this problem that you're working on?
Toomas Hendrik Ilves: Well, I've talked to them with some mixed results. One of the problems was, I think, there was a case of real denial. Some of it may be purposeful, but some of it just like, "We couldn't. We're the good guys. We're connecting the world. We can't be doing these bad things." I mean, this gets into a whole other problem, which I think is one of the things I'm trying to write, or should be writing, which is on the 60th anniversary of C.P. Snow's classic article, "The Two Cultures," which was about basically the university when it was published in 1959, and which was like the science people couldn't care less about the humanities people, and the humanities people really didn't understand a thing about what the science people were Well, now, when he wrote the essay, technology did not impinge upon our lives. I mean, you could watch television, but it couldn't watch you. And your telephone was plugged in the wall, and if you left the room, it didn't know where you are. Whereas, you know, I mean, just reading again today about how someone steps into a store and, you know, a couple of hours later starts getting ads from that store, right? I guess this whole surveillance thing is going to be, I think—
John Markoff: Zuboff calls it surveillance capitalism. Yes, I know.
Toomas Hendrik Ilves: I mean, well, I think basically this is the difference between Silicon Valley and China, is that both track everything you do. But one of them feeds it into the government and the other one doesn't. But in terms of tracking you, I'm not sure it's that different. In terms of basically spying on what you're doing. In one society, there are political consequences for doing things, and with the social credit systems coming in, fairly dire ones. But in terms of the surveillance side of stuff, I mean, they're both very good at it. I mean, there are too many cases of manipulation of— that were not possible before. If you go back to the pre-digital era, okay, you didn't have dark ads. Everyone saw the same ads because they were broadcast, and if something was kind of fishy, you would immediately have an outcry about, you know, this is immoral. But, you know, sort of— who was that? Willie Horton.
John Markoff: Willie Horton and the Swiftboaters were early examples of broadcast disinformation.
Toomas Hendrik Ilves: But they also met with public uproar. Yeah, now you don't even see it. You don't know it, you don't know anything about it. And so, I mean, that's an example of the kinds of things we face today. Psychologically, I think that fake news gets spread 7 times more than real news, because real news is boring. It's boring. I mean, this is like the man bites dog, don't Dog bites man. I mean, a position paper by a reasonable candidate, who's going to read that?
John Markoff: Where will you come down with respect to C.P. Snow? What do you want to add to that economy?
Toomas Hendrik Ilves: On C.P. Snow, it's— I mean, really come down on both sides, which is that looking at the performance of the various parliamentary bodies that have investigated this, both the US Senate but also the British Parliament and the EU, we really need to up the level of basic science knowledge on the part of our legislators in the current era. It just doesn't— I mean, my personal story was when I went to give a talk at the European Parliament in 2004. 2014, when I went to introduce what is sort of for the first time Estonia's plan for its presidency, which was to create a digital single market. Because in Europe you have single market for goods and basic services, but when it comes to digital services, it doesn't exist, which— but anyways, I went to give this talk and there were about 40 members of parliament there listening to strictly voluntary. And so just to sort of make my point, I pull out my iPhone and I said, well, this is— you all have one of these, everyone has one of these. And I said, well, you know, Moore's Law says the power of a chip doubles every year and a half. Your next election is in 4 and a half years. That's 3 iterations of Moore's Law. That means that this thing will probably look the same, but it'll be 2 to the 3rd times more powerful. From the audience, I heard, "What is 2 to the 3rd?" We're talking about a self-selected group of people interested in technology. That's on the one side. Or you can just look at the hearings that took place in the Senate. The Facebook hearings? Yeah. On the other hand, all the things that are coming out in the past 2 years, but surveillance, about selling— Cambridge Analytica debacle, the unwillingness to think about the ethical implications of what you are doing to make a lot, a lot, a lot, a lot of money is appalling.
John Markoff: That spread of that technology to basically the developed world over a very very compressed timeframe, happened in the space of a decade. You come here, you know, if you had been living in Europe and you come to Silicon Valley where these companies are headquartered, did anything strike you? I mean, had the world caught up?
Toomas Hendrik Ilves: Did this seem culturally different in any meaningful way? Well, the first thing that strikes me, and strikes me every time I come back from Europe, is how could this be? I mean, Palo Alto is a town with not one but two Tesla dealerships, two Apple Stores, and even in formerly communist Eastern Europe you don't drive on roads as bad as the ones here. So that already said something. There must be something. Something's kind of odd. And then when you find out, as I found unfortunately out very quickly, the rents that you pay to live here with bad roads. You just go, "How is this possible?" I mean, I was just in Estonia again, and I'm just like, it's— we were a commie country, and now it's, you know, the roads are wonderful, and it's not just in the cities. I drive around and, you know, everything works.
John Markoff: So is there a cultural commitment to infrastructure that we lack in this country? I'm thinking about this in the context of the fact that all of a sudden the debate America is about this thing called socialism, whatever that is.
Toomas Hendrik Ilves: But I don't know what they're— I mean, Estonia has not had a— I mean, Estonia has had coalitions that have included the Social Democrats, but I mean, we're far to the right of even the United States in many ways on economic policy. In my country, and people say, well, how do people trust the government? Well, they don't think of it as government, they think of it as That's what the state is, where those jerks that are in the Prime Minister and those ministers, like, they're— but this thing works, right? But anyway, yes, there are huge differences and that has to do— this is where, for me, the kind of thing that really strikes me is, you know, sitting here and you realize that in a 12-mile radius you have the headquarters of Tesla, Apple, Google, Facebook, Palantir, and probably a whole bunch of other billion-dollar companies. And they do wonderful things. What all of the— I mean, all the things that they have enabled really are fantastic. You know, Hewlett-Packard, plus all the financing over on Sand Hill Road, great. On the other hand, people spend a day, two, three at the Department of Motor Vehicles.
John Markoff: What explains our culture? Do you have any idea why we have this cultural or cultural-technical inertia? Because you guys, I mean, you guys were behind us at one point, but it wasn't like— is it one of those things where because you didn't have the wired infrastructure, you could jump to the wireless infrastructure more quickly, or is it something else?
Toomas Hendrik Ilves: Well, I mean, in our case, I think it was political will because we're far ahead of I mean, Europe as well.
John Markoff: So that comes to vision and leadership. Somebody actually said the path goes this way.
Toomas Hendrik Ilves: Yeah, well, I mean, yeah, I said that 25 years ago and I was belittled and laughed at for at least 7 of those years. So that was, I mean, that's different. No, but I, you don't have to get, I mean, you don't have to be my country, which is I mean, it's such an outlier. And if people listening want to read about it, I would look at the December 17th, 2017 New Yorker piece called "Estonia: The Digital Republic," which is probably the best written thing about it, and written by someone who was a skeptic, went there, came back, and was sort of— But what is it? Well, first of all, lack of attention paid to infrastructure. Of any kind for, I don't know, since, for 30 years, 35 years. Lack of attention to public services that can be made so much easier and better in a digital era. And sort of clumsy, not very well thought out solutions for those things that are digitized.
John Markoff: [Speaker:DC] And so now you're involved in various projects on campus that are responses, like the Spotly Institute has a group and—
Toomas Hendrik Ilves: I'm involved in so many things I can barely keep track of them, but okay, there's the project here at Stanford which is only beginning about how do we address these issues. It's more focused on the United States, but I'm on the chapter 8 of it, which deals with my sort of constant line for the past several years, which is that we need very strong international cooperation. The line of argument there is that basically you have 28 countries in Europe, plus Canada and the United States. All of those countries— there's not a single country that has not been subject to one kind of attack or another. Some more, some less, but it's clear that they all have been subjected to some kind of attack. What kind of cooperation do we have? Zero. I mean, it's—
John Markoff: So you have to build new structures.
Toomas Hendrik Ilves: Yes, I think what you need is either using existing structures, which in one case would be NATO, but NATO is loath to do these. They have a center for cyber in Estonia and a center for strategic communication dealing with propaganda in Latvia. But they're limited in their mandates. That is, they will address issues if the O in NATO, the organization, is attacked. So they'll do brilliant work on, say, disinformation about NATO in the Baltic countries. But there are all kinds of other kinds of disinformation which comes out, which may be sort of targeting a NATO ally, but because it's not specifically against NATO, then it's not within the mandate. In the European Union, the response has been pitiful. And I wouldn't say the U.S. has been much better, in that most of the work that you see done today on dealing with the range of these activities, from bots to disinformation fake news is done by NGOs. Some of them are very good NGOs. I mean, probably DFR Lab at the Atlantic Council and Bellingcat are superb. In fact, Bellingcat probably— open source intelligence is a whole new game after Bellingcat. Before that, people weren't doing that. But if you look at it, you see You have all of these organizations that are doing it, whereas governments are doing almost nothing. And you also see that at least some governments— won't mention any— but don't even want to address the issue. Though you might say that where the attack, the approach to destabilizing, interfering, whatever you want to call it, was successful, the winner of that result is clearly not interested in pursuing this. Not only in the US but also in the UK, you really see this— there's this huge reticence even to address the issue. In fact, basically, the newspaper The Guardian and one journalist, Carol Callowater, who's done all the work exposing how much of this dirty stuff was going on. And okay, there's a parliamentary committee that came out with a report last week, after 2 years after Brexit. Same thing in the United States. There is, instead of boosting and financing more of this, sort of at the government level, you had the White House got rid of its cyber department, now it's back. The State Department dismantled its anti-cyber associated unit and now I guess they're reestablishing it. Whereas France, on the other hand, seems to— where in fact also faced a rather brutal attack in terms of hacking the servers of the Emmanuel Macron, came out very tough because they didn't win, whereas I imagine if Marine Le Pen had won, there would have been "oh no, nothing." Why do you think governments are slow or hesitant to take the reins of some of these issues? I think a lack of knowledge. I mean, if you— I mean, that's why I mentioned this case about, you know, what's 2 to the 3rd? I mean, it's across the world. Basically, your people in government have not really— or in policymaking decisions, you have a lot of people who are smart in government, but they're not making policy. They may understand, but they're not in Congress or Parliament, and they're rarely ministers or secretaries. Occasionally they are, but— so they don't understand the full extent of the threat. I think in the Obama administration, there were people who did, but even there, you didn't see a willingness to actually take major steps in that regard. That's coming out now. Even there, from what I understand from talking to people, that they were worried about Russian meddling, but not of the kind and of the extent to the extent that has become clear. And now there seems to be a complete lack of interest in pursuing these issues at all.
Narrator: If someone came to you and waved a magic wand and you got unlimited staff and budget, what would you get to work on right away?
Toomas Hendrik Ilves: What initiatives? What would you start? I would actually digitize a state in the United States and make it work. And then every other state would see, oh, it works. I mean, you save huge amounts of money. Estonia saves basically 2% of GDP or gets 2% of GDP from simply the time savings. And then that's what I would do. I mean, just because I've done it and I think it would be so cool for someone in the US to do it, which would like be transformative. If the US does it, does it and Silicon Valley sees it, then I think that would be a hugely beneficial thing.
John Markoff: Have you thought about what the first state— is there an ideal state?
Toomas Hendrik Ilves: Well, you need— I would say North Dakota or something where it's easy to— Manageable. Right. I mean, just because it has to be small in the beginning because, I mean, people say, well, what Estonia did, you're a small country, is not scalable. Technologically, there's no problem. But what is difficult to scale is political will. You find that in small societies, smaller countries, you kind of can get to a consensus to do something, and so the countries that have adopted this have all been smaller countries.
John Markoff: Tomas, thanks for chatting with us today. And for our listeners, take a look at the show notes for this episode for links to some of the topics and organizations we've discussed.