Presidential elections HACKED? + Digital Social Norms –– Talking Tech w/ Joe Hall & Kate Klonick
The US Presidential election is less than two months away, and you’ve undoubtedly heard questions about cybersecurity in the lead up to the election (DNC, RNC hacks, etc.). Our resident voting and election cybersecurity expert covers all the issues that could come up this November, and what we should expect.<br><br>Brian also chatted with journalist Kate Klonick about her recent article with Slate on what really governs online speech (hint – not the 1st Amendment). She breaks it down for us.<br><br>Additional info from Joe on online voting cybersecurity: <br>*CNN – http://bit.ly/JoePutinVoting<br>*NPR – http://bit.ly/JoeProtectVote<br><br>Kate is a Resident Fellow at the Information Society Project at Yale and a PhD candidate at Yale Law. Her recent article can be found here: http://bit.ly/KKonlinespeech.<br><br>Attribution: sounds used from Psykophobia, Taira Komori, BenKoning, Zabuhailo, bloomypetal, guitarguy1985, bmusic92, and offthesky of freesound.org.
Top Keywords
- election 0.013
- norms 0.011
- platforms 0.011
- online 0.008
- internet 0.008
- facebook 0.006
- voter registration 0.006
- cybersecurity 0.006
- voting 0.005
- voter 0.005
- registration 0.005
- twitter 0.005
Transcript
Welcome to Tech Talk. Bye. C. D.
Welcome to CDT's tech talk where we dish on tech and Internet policy while also explaining what these policies mean to our daily lives. I'm Brian Wasilowski and it's time to talk tech. We have a presidential election right around the corner and the issue of cybersecurity has never been more central. Whether it's attention grabbing hacks of political parties or the specter of a foreign government tampering with election results, everyone is talking about cybersecurity. Our chief technologist will join us to talk about what concerns are legit and which ones are not. And we'll be talking about free expression online and how in reality, it's not the first amendment that governs how social media platforms manage speech. The US presidential election is less than two months away. For many, that is an incredible relief. This is this election cycle has been unique in many ways to say the least. And regardless of your political leanings, you've undoubtedly heard questions about cybersecurity in the lead up to the election. This includes the hacks of the DNC and d triple c and suggestions that, in fact, this election could somehow be rigged. Well, CDT's phones have been ringing off the hook on this issue because we happen to have one of the smartest people around on voting and election cybersecurity our very own chief technologist Joe Hall and since he has done many a media interview already I thought it was time to get him on tech talk and share some of his insights with our listeners. Welcome, Joe. Thanks. So first, how vulnerable are we to any sort of hacking or tampering or breaches when it comes to this upcoming election?
Sure. So you're not gonna like the answer, which is we don't really know how vulnerable we are. And that's part of the problem is is that the voting machines and equipment that we use now are ten to fifteen year old computers. So they they are not designed to sort of survive in the modern cyber environment if you want to call it that. The other problem is these things don't tend to keep the kinds of records you would want to be able to detect any malfeasance or to even be able to recover from those things, which is why I'm particularly glad something like 70 to 75%
of voters are gonna be voting on machines that create or keep a paper trail of some sort, something that's independent of the software. Yeah. I've noticed that in DC every time I vote regardless of what the election is whether it's a presidential year or not, there is a printout component. Even if you go to a computer, enter your thing, it prints out something and then you scan it in. So that's a that's a good thing? Absolutely. In fact, we call that and the technical term for this is called software independence and the idea is that
no bug in the software could change the outcome of the thing you're trying to do. In this case, vote, you know, elect the people who would actually run our government. And so what you can do is you can compare the paper, the results from the paper to the electronic results as a way of keeping honest. But you have to make sure you keep that paper real safe and secure. And if you can do that, then you at least know that the no errors in the software, errors or malicious activity in the software would change, the actual outcome of of an election, which is
a good thing. So to kind of build on that a little bit, there's a bunch of components when you think about voting, you know, the steps to it. You have the voter registration databases, the voting machines.
Is there anything that is is particularly vulnerable? I know you said we don't know know but. Oh, so we definitely can rank them. For example, the things that are gonna be more networked, so voter registration databases for example. You know you need to be able to register online. That's one of the few things that we think is a really good idea to provide over the Internet in elections is the ability to register to vote. The problem is is because they're online, they provide pretty juicy targets for either, as has been alleged, Russia or other kinds of nation states that might want to do certain kinds of manipulation. But more importantly, hacktivists and other kind of sort of people just looking for the lulls as the the hackers would say, something fun to do. Unfortunately, we've already seen in the state of Arizona and Illinois their voter registration databases being attacked. And the FBI has purportedly linked those to things that look like Russia. But then again, we can never tell if someone's, you know, trying to pretend like they're Russia and doing these kinds of things. But voter registration database, I think, are the ones we're most worried about right now. Interesting. So you mentioned, you know, of course, Arizona,
one other state. It's important to remember that voting is done at a state by state level. So there's different systems in every state, I would assume, at least to some extent. Are there any best practices that you've seen in certain states when it comes to cybersecurity? Are any states doing a really great job on this? Yeah. So, every state does it a little differently.
Some of the best best practices are things like like create backups of your voter registration database. If some nation state hacker were to come in and drop 5% of your, Democrat or Republican voters, you're gonna wanna be able to see that happen and recover from that because the last thing you want on election day are millions of people casting provisional ballots, which are things that take a lot more time. And, so so making backups and checking your backups and being generally vigilant. And it's it's it's a weird thing to say, but being aware and understanding that you may be a target of an attack and you may not be best prepared for that attack, but at least keeping your eyes open and asking questions about, should this have happened? You know, this thing seems anomalous. You know, if that if things like that happen, there's plenty of people that can help you answer those questions. And that's why experts like me are gonna be available on election day to to answer calls from folks. I'll I will yeah. You'll be a busy man on election day. I would imagine that these issues are relatively new to a lot of people that work elections. I mean, it's, you know, it's it's a complicated process to begin with, and now you're adding this layer of cybersecurity, which, I mean, we see in a lot of different settings that it's it's a new skill set that workers have to have in different fields. It's so true. One of my life goals is what I call building a tiny technologist within all of you and trying to make the case that you gotta know a little bit about how all the tech works. And you're right. Election officials aren't folks that are trained to defend systems in sort of a cybersecurity kind of paradigm. And because of that, we need to sort of focus on building that into the job of an election administrator. And no one ever, when they're a kid in first grade says, I wanna grow up to be administer elections. You know? They may say, I wanna be president, which is a really important part of the democracy. But, man, the people who run the machinery of our democracy are extremely important. They rarely get a day off six weeks before an election, and they're gonna have even more on their plate going forward. So we need to support that and make sure they have the resources and capabilities they need to defend our democracy in the face of these kinds of new threats. That's a good point. So you mentioned voter registration online, and this is probably a good time to
suggest to everyone listening. If you haven't done so already, register to vote is an important thing. What about Internet voting? So actually voting online, is that something that you ever see becoming a reality? OMG. Internet voting is just
a really bad idea. It's a bad idea right now, specifically because the state of the Internet. Internet's just not a very secure network. It's not you have to do quite a bit to keep yourself secure while you're on the Internet as you guys can see when you're surfing around and trying to make sure you don't get phished or malware. And clients, you know, the, your mobile devices, your desktops, your laptops, and the servers that run this stuff, those are still pretty insecure. And so all of that, you know, we already have problems with Internet and and devices being very insecure. But add government elections on top of that and man, you do not have a recipe for much good happening. I am sort of a little different in the expert community in that I feel we have to find a way to do this that if you want to have colonies in the Marianas Trench or on the moon or something like that, you're going to need to have democratic decision making without having to transfer physical matter, without having to send a rocket or a balloon from a deep sea outpost or something like that full of ballots, right? And so we need to figure out how to do this. I think it's the question is not when can we vote on the Internet. It's more what would the Internet and devices need to look like to actually give us the confidence that we could do it safely on the Internet. And it's it's gonna be probably twenty or thirty years before we're there. Interesting. And, you know, you have to also wonder
even before we go to Mars, just we tend to have fairly low voter turnout in The US. Some countries do much, much better on that if, you know, you don't really actually have to go to a place. You can see the benefits, but, certainly, I mean, you're very wise to point out the risks there. So last question for you. Election day, November 9, register to vote and go out and vote. We wake up or I wake up to the eighth. I'm sorry. Right? So ninth is the actual next day. Eighth is the last day you could ever have a presidential election. Oh, really? Interesting. Good tidbit. So the ninth, the next day after we've all voted, how are we gonna feel about the results? Should we feel as though these are legit and be confident in what we're hearing? Well, you should all, be very patient.
It takes a long time to count votes. And despite the fact that we are using electronic equipment, people still wanna know at, like, 5PM Eastern time on election day. Projection. You know, who who who won? You know, and that that seems to be, you know, it's still something that irks me because, you know, there's a 30 period after the election where they are checking their math. You know, it takes a long time to check the math. But but I do think that things to look out for on election day or the day after are there are will be allegations of bad things have having happened. There is every election, and, those of us who are deeply embedded embedded in this know equipment fails. There can be, you know, like like Hurricane Sandy. There can be natural disasters that happen. And so, effectively, those kinds of things are are things to look for. I think one of the things that I would be very concerned about is if we actually have, legitimate evidence that someone was able to manipulate the vote. We're a little bit more worried about disruption, things that cause chaos because those are easier to do. It's always easier cause chaos than to do something very specifically subtle. But in those cases, we'll we're gonna really, need when when when, an allegation of actually vote flipping has been made, we're gonna need to look at those and and assess the evidence. And it's unclear if we'll know one way or the other if it has been hacked. I'm hopeful that we'll be able to prove beyond a shadow of a doubt that that we know that certain ballots, especially the Internet voting that is happening for military and overseas voters, that those can be quarantined and put in a separate place. But but those are the things I'll be I'll be looking for is, you know, how legitimate is the evidence and and are there other explanations for that kind of Well, let's hope for a chaos free
November the next day because we've certainly had plenty of chaos in this in this cycle. Thanks so much for joining us, Joe. Thank you. In The United States, we enjoy the right to free speech, obviously, and can tweet or post anything online we want. Right? Well, no, not really. The First Amendment protects us from the government restricting or censoring our speech, but it certainly does not apply to a company or individual restricting speech. Yet, we think of the Internet as a great empowering force that allows us to freely express ourselves and reach audiences worldwide. If the First Amendment does not apply to online platforms, can this be reality or possibly be true? Our guest today, Kate Klonick, explored this question and more in her recent article for Slate. She's a resident fellow at the Information Society Project at Yale and a PhD candidate at Yale Law. As I would say, an all around smarty. Welcome to Tech Talk, Kate. Thanks so much for having me. So if we shouldn't be thinking about the first amendment when it comes to online speech, especially on these social media platforms, what is we should be thinking about or talking about?
I think that the best way to conceive of these things is actually these platforms are forms of government. There's lots of false narratives about what these platforms are doing, and I think that most recently you kind of saw this with the napalm girl photo. There was a lot of chatter about, this was an editorial decision done by human content moderators of these platforms to take down this photo. And I think that gets it really wrong. I this is actually very much an intricate, procedural system that these platforms have developed to govern and decide what we see every morning when we get our coffee and, what we don't wanna see every morning when we get our coffee. So
some people would argue that services like well, or platforms like Google and Facebook and Twitter are really more public services now that maybe, yes, they used they still are companies, but at the same time, they're so ubiquitous that they are kind of public services. Or even at a minimum that if you can't engage on these services in a the way you want, you're you're not able to fully engage in society. Does that maybe make the case that there should be some sort of, like, federal government role, or don't you think so?
Yeah. So I I think that these platforms already do think of themselves as governing. I think that that's, to a certain extent, even if it's privately. But I think that as to the the question of does the does this create a kind of government role for these platforms? I think that basically what's different about the government okay. I just kinda fudge this. I don't know. Go ahead. So I think that that platforms like Google, Facebook, and Twitter already think of themselves as platforms. But I don't know if I'd go so far as to say that they're public services. They're definitely still private companies. But they have they're they're doing things that are starting to have such a large effect over so many people that they're changing how we think about them, in a normative way. And so this is constantly evolving, on the Internet right now, and we have to consider
exactly what we want to interact with on the Internet and what we expect from the Internet, and these platforms are in charge of that. So let's build on that a little bit. I mean, your article, you talked about setting norms for how we behave online and how this is more of a a normative process. What do you think some of these norms should be? Or at least what should be the questions that we're asking when we're thinking about developing and establishing norms for online the online
world? Oh, that's such a good question. So I don't think you we get to decide what the norms should be. I think that that's actually something that is organically happening all the time and constantly changing. I think that's just the nature of norms. And I think that what you can see now is that, we have certain types of norms. I think that the thing to remember is a lot of this is so context dependent. So for example, we don't want to see child pornography in our Facebook feed, in the middle of the day. But we do want to see something like napalm girl. And so how do you create a system that takes that into account and also takes into account all of the norms that are happening all over the world? So American norms are very different than, the norms in India or Sweden. And the norms in Alaska even are different than the norms in Tennessee. So these types of things are all happening online very, very quickly. It's kind of amazing that some of these platforms have developed systems that are as accurate as they are in showing us what we what we do and don't,
get to see online every day. Yeah. No. That's a really good point when you think about the the digital world and the the amount of content you engage with. And sure, there's a it seems to be an agreement there needs to be that the the norms are being established, and sometimes people want more rules. But it's pretty amazing what some of these companies do, especially considering how many users they have. Oftentimes, far more users than a country needs to govern in any sort of way. Oh, absolutely. Oh,
absolutely. And so Facebook, Twitter, or YouTube, they don't have different policies for what stays up and goes down based on where someone is posting, unless it's something very specific. For instance, if it goes against the country's as policies that, that there's, you know, that you sell, Hitler paraphernalia, then you can't see you can't see that while you're in Germany, for example, or France. So there's there's those types of things but, besides that, it's pretty much the same community standards as long as it's not blatantly legal. They have forced the same community standards everywhere and so yes, they basically figured out ways to adjust, those community standards and enforce them, across across boundaries and across, and across borders. It's pretty it's pretty remarkable.
Yeah. It is. So let's go a little beyond some of the the community standards and norms. There are actually terms of services that, these platforms have. And, you know, in your article, you noted that Facebook is is growing in terms of users. Twitter has kind of stagnated and, you know, kind of at least alluded to the fact that you think that terms of services are the way their terms of service play a role in this. What do you think companies are getting right, and what do you think some companies are struggling with when it comes to setting their terms of services?
Yeah. So, I mean, absolutely. And this is just, I mean, to a certain sense, this is just conjecture. You can never kind of know completely why people are leaving Twitter or, you know, going to Facebook. And, but I do think that there seems to be there seems to be, an issue, with how people expect to engage with a platform and expect to be able to engage online. And so at a place like Twitter, if you are going online and getting harassed every day and they're not doing anything about it and you feel very frustrated, why would you keep going going on Twitter? You would just go to a place that is maybe not as public, but you can go to a place like Facebook and feel like you have more responsiveness. So I think that the two things that maybe are, or maybe so helpful in both of these is that, how flexible their policies are, how open they are to kind of an appeals process. If you have something taken down and you don't think it should be taken down, and and then finally, just kind of, looking at how responsive they are to their user group and changing things. I do think that, what is the one thing I will say that's really different about Twitter and Facebook is that they're just different they're just serving different functions. And that has evolved more and more in the last, I would say, five years. YouTube, Twitter, and Facebook are 60% of Americans get their news and go talk online. And all of those services, while they have overlapping features, also have very different roles in how people expect to be treated on each of those sites, where they how they talk on those sites. And, that's, again, part of the not to keep beating this normative drum. But, like, that's part of the that's part of how we're changing norms online and how these things are evolving. The more flexible these platforms can be to those things changing and being circumspect about it,
the better their platform, I think, will be. Yeah. That's a very good point. I mean, I think if everyone just kinda thinks about how they use the different, social media sites that they're on, I mean, my use on all of them has kinda changed. I I've seen my profile on Twitter shift to a much more professional oriented one, although then people tell me I'm boring. And, you know, on Facebook, I've created a much more closed community and have been, you know, stricter on who I curate. And then LinkedIn, gosh, I'm like, you send me a request, you're in. Although someone told me that's not the way to do that either. But definitely, it's, you know, norms are being set. People are still learning how to use these. And it's to me, it's often amazing how creative people are with how they engage on these platforms and how they use them and the types of communities that form that probably I wouldn't predict, but certainly I'm guessing even the platforms themselves wouldn't predict.
Oh, absolutely. But it's it's just it's kind of it's, it's very much just, you know, IRL, like like, living situation laying out online. I mean, it's you know, you wouldn't like, my I have good friends who would come over, and I'd be in my pajamas and, you know, tell, tell funny stories with, and I wouldn't ever do that with my editor, a magazine. Right. And so, and so but but it's hard what's what's so interesting about the Internet is how we're figuring out how to curate how we're perceived by others and then, and then what we when those lines all disappear. So I think that, I think that you're hitting on exactly right. And there's a really good question of if one app can can do all of it, if one platform can do all of it and figure out how to let you silo it without a lot of work. Or if you're just gonna do exactly what you just described, which was use LinkedIn for one thing, use Twitter for another thing, use Facebook for something totally private. And, you know, it's difficult. It's hard. And because not everyone does the same thing as you do. So sometimes you have to find people on Facebook that you thought were just work people. So
A lot of norms that still need to be set. Okay. Before I let you go, tell me about the other research you're doing. I mean, as I said, you're an all around smarty. So what are the other issues that you're working on, that you're finding fascinating right now? Yeah. In the past, I just published a paper, about online shaming. And so that,
that has been a focus of mine. But I'm really working right now on putting out this paper that, I've been researching for the last nine months about content moderation on these platforms and kind of giving legal history and a procedural account of how these platforms make these decisions. And I hope that that's gonna shed a lot of light on a lot of the questions that we're talking about and a lot of the the issues that keep coming up in the news today. So Yeah. No. That sounds issues that keep coming up in the news today. So Yeah. No. That sounds fascinating. I look forward to reading that paper. I've read,
some that have touched on that a bit. And it's but it always strikes me is that you always have to remember that there's a human engaged in this this process, and some of the best moderation systems do have a human element to it still, which I think a lot of people think it's just the algorithms that kind of take care of all this. But in most cases, it's a it's a combination of things. And at a minimum, you know, it's humans creating those algorithms. So lots of fascinating stuff there.
Yeah. Absolutely. I think that that you that's exactly right. It's just an interplay between humans and algorithms and a little bit of both. So yeah. Exactly.
Awesome work. Thanks so much for joining Tech Talk, Kate. Yeah. Thanks for having me. That's it for this episode of Tech Talk. Be sure to visit CDT's website, cdt.org for the latest updates on our work around free expression and cybersecurity. And no doubt, you'll see Joe in the news regularly in the coming weeks. I'm Brian Wasilowski. Thanks for listening.
