Yes, Our Elections Are Secure: Ben Adida

December 12th, 2024

”Election security is in truth getting better over time.”

We discuss election integrity and voting machines, the amazing work of election workers, and the logistics of election audits.

Ben Adida is the Executive Director of Voting Works, the only non-profit voting system vendor in the U.S. Their mission is to make election technology everyone can trust through transparency, simplicity, and demonstrable security.

Your civic action toolkit recommendations from Ben are:

1. Talk to your town or county clerk and offer to help.

2. Be a poll worker for a day!

Let’s connect! Follow Future Hindsight on Instagram: 

https://www.instagram.com/futurehindsightpod/

Discover new ways to #BetheSpark: 

https://www.futurehindsight.com/spark

Follow Mila on X:

https://x.com/milaatmos

Follow Ben on X:

https://x.com/benadida

Sponsors:

Need a gift idea? Head over to Masterclass.com/HOPEFUL for the current offer. Thanks to MasterClass for supporting Future Hindsight!

Thanks to SelectQuote for supporting Future Hindsight! Go to SelectQuote.com/hopeful now!

Follow Future Hindsight on Instagram:

https://www.instagram.com/futurehindsightpod/

Early episodes for Patreon supporters:

https://patreon.com/futurehindsight

Credits:

Host: Mila Atmos 

Guest: Ben Adida

Executive Producer: Mila Atmos

Producer: Zack Travis

• Transcript

  Ben Adida Transcript

  Mila Atmos: [00:00:00] Thanks to SelectQuote for supporting Future Hindsight. SelectQuote is one of America's leading insurance brokers, with nearly 40 years of experience helping over 2 million customers find over $700 billion in coverage.

  And thanks to Masterclass for supporting our show. For a gift that's always on time and lasts a lifetime, you cannot do better than Masterclass. Masterclass always has great offers during the holidays, sometimes up to as much as 50% off. Head over to Masterclass.com/hopeful for the current offer now.

  Oh, and before we start the show, I have to remind you that it's the season of giving. If you want to think outside the box this year, don't worry, we've got you. Once you're done with the episode, head over to Futurehindsight.com to check out our brand new merch. We've got something for everyone. That's futurehindsight.com.

  Mila Atmos: [00:01:00] Welcome to Future Hindsight, a podcast on a mission to spark civic action. I'm your host, Mila Atmos. I'm a global citizen based in New York City, and I'm deeply curious about the way our society works. So each week, I bring you conversations to cut through the confusion around today's most important civic issues and share clear, actionable ways for us to build a brighter future together. After all, democracy is not a spectator sport. Tomorrow starts right now.

  Democracy is a concept of government by the people, for the people. And a large part of that, of course, is voting. Now, voting is only one aspect of democracy, but it all starts there. It's the foundation. Therefore, we need to be able to trust the results of an election. Election integrity is essential to democracy.

  Our guest is Ben Adida. He's the executive director of Voting Works and a world renowned election security expert who received his PhD from MIT's Cryptography and Information Security Group, where he focused on election security. Voting works is the only nonprofit voting system vendor in the U.S., and their mission is to make election technology everyone can trust through transparency, simplicity, and demonstrable security. Ben is also the host of How Voting Works, a limited series podcast on the mechanics and technology of how we vote in the United States.

  Welcome, Ben. Thank you for joining us.
  Ben Adida: [00:02:36] It's so great to be here. Thanks for having me.

  Mila Atmos: [00:02:39] So after years of sowing doubt about the integrity and security of elections, the big lie January 6th, the 60 plus audits across the states, voter intimidation, etc., etc. we can agree that passions run very high on the subject of free and fair elections. But on a very basic level, what exactly is election integrity and why should we care?

  Ben Adida: [00:03:05] Oh, that's a deep question. I could go on for a while there. Let's see. At the end of the day, if we're going to have a democracy, it depends on everybody knowing and feeling that their voice was heard. Right. Knowing that the choice they expressed made it into the tally. And there's a number of things that go into that. The first one is everyone should have a chance to vote. Everyone who's eligible should have a chance to vote. The second one that's a little less well understood is everyone should really have a secret ballot. They should have the ability to vote their conscience without anybody else finding out what they voted for, so that they can vote their conscience. Right. And then they should have confidence that their vote made it into the tally. So election integrity at the end of the day is about all citizens knowing and feeling that they're actually in a functioning democracy.

  Mila Atmos: [00:03:58] That's very well put. I love that you really connected it back to democracy as opposed to leaving it as a technical answer. So I'm curious about how voting machines fit into this picture. What's the function of voting machines in delivering election integrity?

  Ben Adida: [00:04:17] So I think one of the questions that everyone should ask, and that a lot of people do ask is, why do we even need voting machines, right? There are a lot of democracies that don't use them, and they seem to do just fine. And so of all the various questions that people have about voting systems, the one that I have the most empathy for is this one. Why do we need voting machines? The answer is surprisingly simple, and so much so that sometimes people don't really believe it. We need voting machines in the US because of our ballots, because the ballots we have are very long,

  much longer than they are in other democracies. So whereas lots of other democracies might have one question or two questions on their ballot, in the United States, we tend to have dozens of questions on our ballots. There's even a case in Texas where there were 92 questions on a particular ballot. So everybody usually thinks of California as the place where there's too many questions. We know propositions, etc., but California is not the only one. There's plenty of states where there are many, many questions. And then in addition to that, you get variations in ballots from voter to voter within the same county, depending on where they live in the county. Right. So we're all voting for the same presidential contest, but we're not all voting for the same school board. And as a result, you have these different ballot styles. You know, two people living across the street from each other might have 80% of the contests they're voting on be the same, and 20% of them be different. So when you look at all the combinatorics of that, the number of contests, the variability between ballots. We don't have a way to count those ballots by hand in any reasonable amount of time, with any reasonable amount of accuracy. So we need machines. Like that's the bottom line. We need machines. And that's where machines come in. They come in to count our very complicated ballots.

  Mila Atmos: [00:06:09] Got it. Even so, it takes a long time to count all the ballots, it sounds like, because California appears to take over a month to count the ballots. And I'm a little bit curious about that. Like, why does it take so long if we have machines?

  Ben Adida: [00:06:25] It's a great question. So it's not a technical reason why it takes a long time in California to count the ballots. Although there might be some technical contributors to the opening of vote by mail envelopes, that might be a little slower in some cases than others. It's mostly a policy decision, which is that in California, every voter has the right to vote by mail. You don't need an excuse. In fact, since Covid in California, every voter receives a vote by mail ballot by default, they can still come in in person and vote in person if they choose to. But if you don't do anything, as long as you're registered to vote, you will get a ballot in the mail and you'll be able to return it. And California has made the policy decision that as long as your ballot is postmarked by Election Day, then it counts. So the mail might take a few days. Right. And then they need to do identity checks on the envelopes and open the envelopes and scan them. And so when you put all of that together, it might take a little while to scan those ballots. Plus, sometimes the signature doesn't match right or it looks like it doesn't match. So there's a cure process where you look at an envelope and you go, hmm, I'm not sure

  that signature matches what we have on record for this voter, but it could be that, you know, as they got older, their signature changed, or it could be that they're Gen Z and they never sign anything. And so as a result, their signature is not consistent, which is actually a thing that we noticed in the last election. So there's in many counties, there's a cure process where the county is going to notify the voter, hey, we have a problem with your ballot. Can you come and make sure that it was actually you? And that's going to take a few days to validate that that ballot was actually correct. And then it's going to get opened and counted. So in a contest that's very tight, and there are a couple of congressional seats in California that are very tight. It actually matters. You have to wait until all of that is done to decide who won the election. What's interesting is that there's plenty of other states they typically don't take as long as California does, but there's plenty of other states that still take a number of days, even sometimes weeks, to really get the final count in to get all those last votes adjudicated. But in most states, it doesn't matter, because you know what the outcome is going to be, because the margin is big enough that those last few votes don't matter. So the public, the public stops paying attention, but the process is still ongoing. Yeah. California, I would say, is a little bit in a class of its own.

  Mila Atmos: [00:08:52] Yes, but all this to say, you know, we were talking about curing votes and making sure everything matches and the races are so tight. Every vote counts. It's always worth repeating that every vote counts.

  Ben Adida: [00:09:05] I think that's the thing. If you ever spend time working in elections, it's the thing you realize when you talk to election administrators across the political spectrum. You know, as deep red or as deep blue as you want to go in America, election administrators are so committed to getting every vote to count, and it's by far the most inspirational aspect of the job. When you see that, you know, in person at the precinct level, at the single vote level, just the amount of care is really incredible.

  Mila Atmos: [00:09:32] Yeah. You know, I actually watched all of the videos on your website with the election officials in Mississippi.

  Ben Adida: [00:09:40] Yeah.

  Mila Atmos: [00:09:40] And I was really touched by how earnest and sincere they were in making sure that every vote is counted. Tell us a little bit about the humans who run elections. And you know, from your experience in being in direct contact with them and the way that the average voter isn't, you know, the average voter goes in votes and leaves.

  Ben Adida: [00:10:01] Yeah. So the thing you should know about most election administrators is they took the job to not be in the public eye to do something of value, of meaning. But for most of them who have been doing this for years and years, this was never a high profile job. This was never a job where they had to talk to the press very often. In fact, they generally don't like talking to the press because that usually means that something went wrong. And so now they have to, you know, to talk to somebody about it. They do the job because they really care about the democratic process. And not only is their Election Day an incredibly long day, but the weeks and months that lead up to an election are incredibly intensive. And then the thing that most people don't realize is elections are always ongoing. In some states, you have occasionally a year out of four without elections. But in most states you have elections every year and oftentimes multiple times a year when you take into account local elections, special elections, general midterms, etc.. So they're constantly working on this and they care about every last detail so much. They're very process oriented. They're very checklist oriented. They're looking at every way in which something goes wrong. One analogy that's often used in the election space is that, you know, think about what it takes to organize a wedding. If you've organized a wedding,

  Mila Atmos: [00:11:19] Yes, I have.

  Ben Adida: [00:11:19] You know what it takes. It's it's a it's a lot of work, right. Now imagine you're doing that. But like a hundred weddings in the same city, at 100 different locations, that's the kind of logistics you're dealing with when you have 100 precincts and every single hour of the day and every precinct has to open at 7:00 or 8:00, whatever it is, when you have the line out the door. The logistics of it are daunting. And so I'm talking in generalities, but the story I always tell at Voting Works, we had the chance to help the state of Georgia with their audits this past election, but also in 2020, when the presidential election in Georgia was really tight. You may remember that Joe Biden won Georgia by 0.29% in 2020, and we were helping Georgia run their post

  election audit, which is a check on the tabulation process. And there was a deadline. All the counties had to report their numbers by I think it was 6 p.m. on a particular day of the week, and there was one county that was off by one ballot. And in an audit, it doesn't matter. Off by one ballot in a county like it's not a big deal in the end game, because you're just checking to see that overall things are accurate, right? That election administrator was adamant about finding that last ballot. So I get a phone call from one county clerk, and the county clerk says, I found the ballot, I found the missing ballot, and I said, it's okay, it doesn't matter. And we're past the deadline, so don't worry about it. You know, everything is okay. And she was adamant. She said, no, I need this ballot counted. And so she had me call the Secretary of State's office in Georgia to ask for special permission to reopen the audit software so she could enter that one additional vote. And that story sticks in my mind because she just cared so much. And this is after running an election that was really tight and audit that had a lot of people watching in the middle of Covid, November 2020. And she really wanted that last vote counted. And, you know, she's the she's the archetype. I see a lot of folks like that around the country who care deeply about counting every vote.

  Mila Atmos: [00:13:22] Yeah, that's really heartwarming. And it really makes me, it really makes me sad that not more people vote, right. When you have people like this, so committed to doing this process and making sure that our votes are counted.

  Mila Atmos: [00:13:37] We'll be right back with Ben Adida in a moment. So stay close. You definitely don't want to miss this episode's civic spark. One small step we can all take to be more empowered and ignite collective change. But first, we want to tell you about our sponsor, SelectQuote. Nothing surprises me anymore, and it seems like every day there's something new to worry about. So much is out of our control. But one thing you can regain control of is your family's financial future with life insurance through Selectquote. SelectQuote is one of America's leading insurance brokers, with nearly 40 years of experience helping over 2 million customers find over $700 billion in coverage since 1985. Other life insurance brokers offer impersonal, one-size-fits-all policies that may cost you more and cover you less. While SelectQuote's licensed insurance agents work for you to tailor life insurance policy for your individual needs in as little as 15 minutes. And have you ever worried about getting coverage with a pre-existing health condition. SelectQuote partners with carriers that provide policies for a variety of health conditions. If you have high blood pressure, no problem. If you have diabetes, that's fine

  too. Even if you have heart disease, SelectQuote partners with carriers that can cover those conditions and others. Or if you don't have any major health issues, they work with carriers that can get you same day coverage with no medical exam required. Head to selectquote.com and a licensed insurance agent will call you right away with the right policy for your life and your budget. SelectQuote. They shop, you save. Get the right life insurance for you for less at selectquote.com/hopeful. Go to selectquote.com/hopeful today to get started. That's selectquote.com/hopeful.

  Mila Atmos: [00:15:42] We're also excited to tell you about Masterclass this holiday season. Don't just give your loved ones something they want, give them something they want to be. With Masterclass, they can finally become a better cook, writer, entrepreneur or artist like they've always wanted. That's the gift of masterclass. As someone who reads a lot of non-fiction books for the podcast, I have really come to enjoy my experience with Masterclass because of how much I learn. I got goosebumps from listening to Madeleine Albright and Condoleezza Rice talk about democracy in America and around the world. Madeleine Albright even said that democracy is not a spectator sport! And maybe it's out of left field for Future Hindsight, but our producer, Zack, is an avid camper and loved Jessie Krebs' course on Wilderness Survival. So you can trust me when I say, with Masterclass, you and your loved ones can learn from the best to become their best in this season of giving. Gift a year of learning with Masterclass. Plus, there's no risk. Every new membership comes with a 30 day money back guarantee, and right now our listeners can receive up to 50% off at masterclass.com/hopeful. Masterclass is the only streaming platform where you can learn and grow, with over 200 of the world's best. That's why Wirecutter calls it an invaluable gift. So give the gift of unlimited learning. Learn from any Masterclass instructor anywhere on a smartphone, computer, Smart TV, or even in audio mode. You know, just like us. The possibilities are endless. Truly help your loved ones learn the art of storytelling with Michael Lewis. Use science to solve problems with Bill Nye, or grow their communication skills with George Stephanopoulos. And the best thing? The classes really make a difference. 88% of members feel that Masterclass made a positive impact on their lives. Masterclass always has great offers during the holidays, sometimes up to as much as 50% off. Head over to masterclass.com/Hopeful for the current offer. That's up to 50% off at masterclass.com/hopeful. Masterclass.com/hopeful. And now let's return to my conversation with Ben Adida.

  Mila Atmos: [00:18:13] So I want to turn to the voting machines. I know that there are multiple types of voting machines. And in the current time, what do you perceive to be the gold standard of voting machines?

  Ben Adida: [00:18:24] So the dominant system, which I think is the gold standard for how we should vote, is that many voters will fill out a paper ballot, a bubble ballot, right? Fill in the bubbles next to the names that they want to, of the candidates they want to vote for, and either scan that ballot in person at a precinct, or send it by mail and have it counted by a batch scanner, basically. And then you complement that with a voting machine for voters who might have a disability or might prefer to use a different style, which is a touch screen or an accessible controller plugged into that touchscreen, or any number of controllers for voters with disabilities, like a sip and puff machine for voters who don't have the use of their hands. And that machine will help them prepare a ballot, print it, verify it, and cast it. And having those two is not only legally required, but actually a really good idea. And there's a lot of discussion around, you know, should the focus be only on one or the other in general? My take of the gold standard is that you give voters the choice. Which one do you want? You shouldn't make assumptions that it's only blind voters who should use the touchscreen machine, because there's lots of mild disabilities or personal voter preferences. You know, one of the common criticisms people have of filling in bubble ballots is if you have arthritis.

  Mila Atmos: [00:19:50] Mhm. It's hard to do.
  Ben Adida: [00:19:51] You're feeling. Yeah. You got to fill in 80 bubbles on a ballot.

  Like that's going to be really painful.
  Mila Atmos: [00:19:56] Yeah. It's tedious.

  Ben Adida: [00:19:57] So the gold standard in my mind. Yeah. Super tedious. Yeah. So the gold standard in my mind is having those two options and letting voters decide. But always, always a paper ballot.

  Mila Atmos: [00:20:05] Always a paper ballot. Well, let's talk about Voting Works. Ben Adida: [00:20:07] Yes.

  Mila Atmos: [00:20:08] Tell us about your origin story. How did you come to be? And why are you in this space?

  Ben Adida: [00:20:13] Yeah. So I've been fascinated by voting for the same reasons you discuss in your introduction, the inspiration of democracy and the importance of of that. And I'm a technologist at heart. I have been a tinkerer since I was a kid. So for me, it was always about finding like, how do I apply my skills and my and my love of technology to something that is a public good. And voting just came almost naturally to me very, very quickly. Even as an undergraduate where I had the chance to work on some interesting voting, cryptography, and some ideas of how do you secure voting systems. And I did that partly as an undergraduate. Then it was the focus of my graduate work for my PhD, and I took a bit of a break from voting, although it wasn't really a full break, I was still involved as a as a hobby, but I went off and did various gigs in technology companies, leading product engineering teams. And around 2018, after more than a decade of doing that, I realized I needed to come back to voting. It was this, you know, I have friends and family who will tell you that, like, you just can't stop talking about it. It's clearly the thing you need to come back to. And I felt like I could use the experience I'd gained building products, leading technology teams to now bring those skills to voting. And so I did that at the end of 2018 with the idea that we could build better voting systems that were more transparent, that were easier to use. I'm not going to say that I predicted the crisis of trust of 2020 because I didn't. You know, it's not like I saw it on the on the horizon. Um, but the ideas that have fed some of the election denialism. I don't agree with them, but I understand that they're reinforced by the lack of transparency and the lack of simplicity in this field of voting systems. It's just these very opaque, hard to understand systems. And if you ask too many questions, you're just told, like, you know, don't worry, it's all okay.

  Mila Atmos: [00:22:09] Yeah. That's not reassuring.

  Ben Adida: [00:22:11] That's not reassuring. So the goal of voting works is can we do it differently? Can we build a voting machine that everyone can trust? And by the way, part of it is coming on a podcast like this and being willing to talk about how it works and the ways that voting systems are imperfect and answering questions which honestly, people in this field don't do enough of. Right? People who are building voting systems

  don't talk about it enough. We should be talking about it. We should be educating the world about how these systems work.

  Mila Atmos: [00:22:36] Right, right. Well, talk about your voting machines because they're the only ones that are open-sourced, which means that it's the only voting system running fully open source software tested to the latest security standards, which I'm now quoting from your website. But why is that important? How are your voting machines different from other machines?

  Ben Adida: [00:22:54] It's a great question. So open source, the reason we think it's important, there's a number of of ways that that's important. One is you just have fewer secrets about how the system is built. Right. If you're wondering about something like, well, how exactly do you decide if a bubble is sufficiently filled, right. You know, is is one check mark enough? Do I have to fill in the full? How does it actually work? Well, you can go and check out our source code. You can see exactly how it works. You can see the tests that we use. Like here's a test of a filled bubble. Here's a test of an unfilled bubble. Oh, okay. And just that level of transparency inspires vastly more confidence because we're just being open about how the system works. Right. You can even see the tests that we've run over years where we encountered a her buggy situation in one election, and then we went and debugged it and tried to figure out why was the ballot not being read correctly. And then we built a test to make sure that in the future we would always read that ballot correctly. So we think of it as a much more honest and open way to develop a system. And we think that it's something that you owe the public when you want to count their votes. Right. That's that's a critical thing. There's another aspect that people don't think about quite as much, but actually is really intuitive. If you walk into an open kitchen restaurant, right, you can see the chef's cooking. You can see exactly what they're doing. It's a lot less likely that one of the chefs is going to take a shortcut and take a knife from the raw to the cooked stand, you know, because everybody's watching, right? We're all humans. We all take shortcuts when the opportunity presents itself. Like, we're not more special than, we're not, we're not more perfectionist than other software engineers. But we know that by doing everything we do in the open, everything's in the open. When I write a line of code for the system and another software engineer, comments on it and pushes back and says maybe it should be done that way. All of that is public. All of that discussion is open for the world to see. By doing the work in public, like an open kitchen restaurant, we're forcing ourselves to a

  higher standard because we know we're being watched all the time. And I think that creates a better product.

  Mila Atmos: [00:24:51] Oh yeah, for sure. I mean, now that you've explained it this way, this makes a lot of sense. I have a question about the paper ballots. You know, having the paper ballot proof after you vote and make sure that it's actually what you voted and the auditing process because you mentioned the auditing just now in Georgia. And this is maybe a simple, maybe dumb question, like what actually happens when people recount and do audits. Give us like a little bit of a detail because I think people are like they're doing the audit, but we don't actually know how are they doing it?

  Ben Adida: [00:25:20] Yeah, it's not a dumb question at all. It's actually a question that more people should ask to understand, like what is this audit about? Right? And there are different kinds of audits. And the ones that we do at voting works are specifically tabulation audits. So what's the idea? The idea is that if the ballots are paper ballots that the voter can look at and see, either because they filled it in or because the computer filled it in for them, but they can look at it. That is the best way we know to make sure that their ballot was recorded according to their preferences, and no one else has to be watching that. Right. It's you in the voting booth with the with the paper ballot, that's it. And then after that, we keep a chain of custody of all those paper ballots. Right. They go in a ballot box, there's a seal on the ballot box. There's a number on that. And we track all those boxes of ballots all the way through. Now, those ballots are being tabulated by a computer, right? That scanner is a computer. And computers can make mistakes. Those mistakes could be honest mistakes. Like there was some dirt on the scanner. And so some bubbles didn't get read as well as other bubbles. Right. Or it could be malicious. Russia hacked the voting machine. Right. And the entire range of that is possible. And so the point of a good audit is to go back to the source, to go back to the paper and check with humans, not machines anymore, that the count is valid based on the paper. Now, we can't do it by looking at all the ballots, because we already know that would take way too long. It would take forever and be super expensive, right? But we can do it by looking at a sample of the ballots, right?

  Mila Atmos: [00:26:53] OK. So you don't count, recount everything. You just recount a sample.

  Ben Adida: [00:26:56] With a caveat.
  Mila Atmos: [00:26:57] Okay. What's the percentage?

  Ben Adida: [00:26:59] Yeah. So it's a great question. What's the percentage? It depends on how tight the margin is.

  Mila Atmos: [00:27:04] Okay.

  Ben Adida: [00:27:04] If somebody wins by 20%, 60 to 40, right. You're not going to have to look at that many ballots randomly selected to see that come out. Right. So so so the larger the margin, the smaller the sample.

  Mila Atmos: [00:27:17] Right.

  Ben Adida: [00:27:17] It's really, really easy to do an audit of an election that's 60 over 40. But if you get an election like Georgia in 2020, that was 0.3% really tight to begin with, then you actually have to look at a lot of ballots to get there. And it is possible if the margin is tight enough, and if there are some errors encountered in the audit, it is possible that you escalate to looking at all the ballots, actually. But the point is, you do that only if it's strictly necessary, right? So you want to do the audit as part of the process of voting. It's just good hygiene. Always do the audit right. And potentially if you're in a state where things are really tight and maybe there's a few errors encountered during the audit, potentially you might end up doing a full hand count, but only if you really need to.

  Mila Atmos: [00:28:00] Right. So in Georgia, how many votes did you recount?

  Ben Adida: [00:28:04] So that's a great question. And there's a little bit of a of interesting in the weeds details. In Georgia in 2020, the math that we used to do the statistical sample was an earlier version of the algorithm for sampling than what we have today. As a result, because the margin was so tight, we would have had to look at 1.7 million ballots out of the 5 million ballots cast. So a third of the ballots, and that's just not efficient. You might as well count all the ballots if you're looking at every third ballot. Just the process of selecting those third ballots is more work than just looking at

  everything. So the 2020 audit in Georgia is one of those rare audits that escalated to a full hand count.

  Mila Atmos: [00:28:46] I see.
  Ben Adida: [00:28:47] To be clear. Just to the presidential contest, right. It's not like we

  were recounting all the other contests. Mila Atmos: [00:28:50] Right, right.

  Ben Adida: [00:28:51] And that took a week and hundreds and thousands of people actually across the state, right to, to to do that process.

  Mila Atmos: [00:28:56] Yes.

  Ben Adida: [00:28:56] It was crazy. What's really interesting is as the statistical process for this gets better, as the techniques for auditing get better, you can actually get higher confidence from less work over time. So if we'd had to rerun that same audit in 2024 with the improved statistical methods, which, by the way, are all published, peer- reviewed. They're not all done by Voting Works. In fact, most of the fundamental research is done by statisticians who don't work for Voting Works. We take that public research and we build it into the software that we have. If you had to redo that audit now, I think we would have been able to look at only 7 or 8% of the ballots, instead of 100% of the ballots to get the same level of confidence. So these audits are getting better, and they can yield higher confidence with less work. And they're a really important part of modern election security. And we're doing them in eight states now across the country.

  Mila Atmos: [00:29:45] Right. Well, so what you're saying is that even though in this environment of doubt, in election integrity, the election is more and more secure in general as time progresses and the technology improves?

  Ben Adida: [00:30:01] I think that's right. You know, we focused on the audits just now, but I think we are, over time, moving to almost every American voter, having a voter verifiable paper ballot. Right. We talked about how the paper ballots are so important.

  That wasn't the case 20 years ago. A lot of voting in the US was electronic in the early 2000. No paper ballot. We moved away from that. So I do think election security is, in truth, getting better over time, right? The thing is, the secrecy of the existing industry kind of overshadows that. And so even though the election security has gotten better, the perception of election security has gotten worse. And in a world where there's broken trust -- and there has been broken trust -- it takes a lot to rebuild that trust. And that's one of the things we are focusing on at voting works is how do we rebuild that trust. And I think extreme transparency is one of the fundamental ways to do that.

  Mila Atmos: [00:30:54] Right, right. Well, I have two questions. Follow ups from there. So the first one is you're an expert on cryptography, so there's no better person to ask about this. Can you tell us about what it would take to actually hack an election? You know, you mentioned earlier, let's say Russia hacks the election. Like what is the theory of how one would be able to steal elections with voting machines?

  Ben Adida: [00:31:21] That's an excellent question. And interestingly, it doesn't come down to a lot of cryptography. It comes down to a lot of system, what we call system security and layers of defense.

  Mila Atmos: [00:31:32] Right. Okay.

  Ben Adida: [00:31:32] And so the fundamental principle of security in practice is that there isn't a single layer of defense that is going to be perfect. Every defense you build is imperfect. Right. And so you build layers with the goal being that for an attacker to penetrate all those layers, it would take a while and it would be very hard and expensive and they would probably get caught. Right. The analogy I give for people my age is the vault in Die Hard, the movie. At the end, where they have to break through the seven doors to get into those bank bonds. Right. Like, that's the way security is done in practice. You build those layers of defense and you, you know, one has to be broken this way. The other one has to be broken another way. Right.

  Mila Atmos: [00:32:11] Right.

  Ben Adida: [00:32:12] In practice, that's how you defend things. So what are the layers of defense in a voting system? The first one is the paper ballot. If you wanted to hack an

  election so that audits wouldn't catch the hack so that recounts wouldn't catch the hack. You would have to somehow stuff paper ballots or change paper ballots in all the precincts in all the counties that you're attacking. And that's, that's already pretty freaking hard, right, to do that. So the paper ballot is a massive security layered defense. What else would you have to do? You would have to potentially install software, malicious software on the tabulators so that they would tabulate the paper differently. Right. Well, those machines are not connected to the internet. That's an important layer of defense. It's not perfect. But it does reduce what we call the attack surface significantly. Right. If you have a machine connected to the internet, then anybody on the internet can try to connect to it and try to attack it. If the machine is never connected to the internet, you have to show up in person. It's a completely different it's a completely different thing. Right. So that's another layer of defense, right? Audits are another layer of defense. Ballot accounting. Election administrators know exactly how many ballots they print. They know exactly how many ballots they give out, and then they count the ballots that remain. This is the thing that most people don't know. At the end of the election, they count the ballots that remain, and those numbers have to add up. The ballots that remain, the ballots that were handed out, the total ballots that were printed. Right. So another layer of defense. So the way you would have to attack an election if you were trying to attack an election, you would have to corrupt all those layers, right?

  Mila Atmos: [00:33:43] I see. Yes, you have to do every single layer in order to actually be successful.

  Ben Adida: [00:33:47] That's right. Whereas for example... Mila Atmos: [00:33:49] that's unlikely. Right?

  Ben Adida: [00:33:50] It's very unlikely. It's very hard. And you're probably going to get caught, right?

  Mila Atmos: [00:33:53] Yes.
  Ben Adida: [00:33:53] Is what it comes down to.

  Mila Atmos: [00:33:54] Right. Right.
  Ben Adida: [00:33:57] Compared to the fully electronic voting machines that existed in

  the early 2000s.
  Mila Atmos: [00:34:00] Which was easier to manipulate. Relatively.

  Ben Adida: [00:34:03] All you would have, much easier. All you would need is a software hack. I mean, all. It would still be tricky, but that would be the only thing you need because there's no paper backup, right?

  Mila Atmos: [00:34:12] Right. You can't prove. That. It's difficult to prove that something happened.

  Ben Adida: [00:34:15] Right. You corrupt the memory of those machines. There's no trace. Right. And that's one of the reasons why security experts have been saying we need those paper backups. It's really, really important.

  Mila Atmos: [00:34:26] Really important. Right. So I've read on X that there's a low grade grumble, you know, by some disgruntled Democrats that the election was stolen this time by Trump because he publicly said he doesn't need your vote. Don't worry, it's going to work out. And of course, the bomb threats from Russia to polling stations in the state of Georgia. So what do you think about the results this year?

  Ben Adida: [00:34:52] I mean, with the admission that I'm a little biased by the fact that Voting Works is helping to run all of these audits, right? And I think we're doing a good job. I think the results of the election are fair. Trump was elected. He won all the swing states. I don't think there's any significant reason to question that. I do think thinking about how you decide to question an election is an interesting question, right? Like it could happen. We could have a hacked election, right? It's not something that could never happen. The way you want to think about it is you need evidence. The name of the game is evidence.

  Mila Atmos: [00:35:23] What is the evidence?

  Ben Adida: [00:35:24] Yeah. What is the evidence that something was hacked? Because on the flip side of that, when you're an election security person, when you're a vendor, when you're an election administrator, what you're doing is you're trying to build up a lot of evidence that things went well. Well, here's my log of voters. Here's my poll book, here's my number of ballots. Here's the audit that I do at the end of the day. Here's, you know, the testing that we do on the machines. Like, is it perfect evidence? No, none of it is ever perfect. But it's a nice, sizable mountain of evidence that things went well. So if you want to come at that and say, I think something went wrong, it behooves you to not just connect weird little dots and say, well, you know, I heard that some of these precincts used Starlink to connect to the internet and to to download their voter lists. And Elon owns Starlink, therefore... Right. Like, that's just not enough. That's not enough to even have a reasonable conversation. Right? You need to bring real evidence to the table. And if you do, then great. Let's investigate. Right. Like we should investigate these things. The thing that folks need to understand is you're never going to get a perfect mathematical formula that proves to you that the election went well. That doesn't exist. What you get is a lot of evidence that it went well, all of it imperfect, but taken together, fairly convincing. And that's going to feel imperfect to some people that's going to feel insufficient.

  Mila Atmos: [00:36:41] Right. But that's what it is.

  Ben Adida: [00:36:43] That's what it is. And you know why? It's because of the secret ballot. Because of the secret ballot, where we cannot have a list of names and who people voted for. If we could do that, if we could have "Ben Adida voted for so-and-so," right. And everybody else. And how they voted. And we could publish that list. We would have a mathematical proof of the outcome. We would know. But that list should never exist.

  Mila Atmos: [00:37:03] Yes.
  Ben Adida: [00:37:03] And because it doesn't,
  Mila Atmos: [00:37:04] Thank God it does not exist, actually.

  Ben Adida: [00:37:05] Right? No. That's really, really bad. Right. That would be very bad for democracy.

  Mila Atmos: [00:37:08] Very bad. Yes.

  Ben Adida: [00:37:09] Because that list doesn't exist, all we have is a lot of evidence of proper chain of custody, of proper accounting. And it's pretty good, right? It's just not a mathematical certainty. And some people will be always upset by that.

  Mila Atmos: [00:37:23] Yes, well, we're talking about trust and about secrecy. Let's talk about the Dominion defamation lawsuit against Fox News, which, of course, Fox News decided to settle for $787.5 million in April of 2023. They admitted -- Fox News admitted that they lied about Dominion Voting Systems committing election fraud. So how has that lawsuit impacted the election process or the voting machine space?

  Ben Adida: [00:37:54] Yeah, the lawsuit itself, I don't think has changed much. But the situation and the accusations around Dominion, which I should be clear, I think are bogus, right. Like, I think those accusations against Dominion were totally bogus.

  Mila Atmos: [00:38:07] I mean, Fox News admitted that it was bogus and paid and paid a fine.

  Ben Adida: [00:38:11] They paid a heavy price for it. That's right, that's right. And I think that's good. I think that, you know, large organizations that spread lies about elections need to be held accountable. I think that's important. Otherwise, you know, we're just incentivizing more lies. Right.

  Mila Atmos: [00:38:24] Yes.

  Ben Adida: [00:38:25] But I think the attacks against Dominion and the attacks against election administrators in particular, because it wasn't just against Dominion. Right. It was a number of election administrators that were attacked. Those did a lot of harm to the election community, to people who, as I mentioned earlier, just were not equipped to deal with these kinds of attacks and defamation. Right. Because it's just not part of their job. They are hunkering down with checklists and trying to follow a process, and

  suddenly they're being told their voting machine is hacked by China or God knows what. Like, you know, this was not their job to deal with this kind of stuff, and it has caused some significant harm. And that's quite upsetting because some folks, a lot of folks have quit, right? They don't want to deal with this, right? They don't want threats to themselves and to their families. Like that was not anything they signed up for. This is why I encourage folks to think about the evidence they have. Right? And in 2020, it was like, what is the actual evidence you have that Dominion was hacked? Right? You can't just hand wave and connect dots. You have to have a real logical argument in 2024. What is the evidence you have that Elon hacked voting machines? You can't just connect these dots and, you know, vaguely imply, oh, that looks a little fishy. Because when you do that, you participate in a system that ultimately leads to really good people working for democracy getting threatened. Even if you're not the one who made that threat, you're participating in that. And don't do that. By all means, learn about voting. Participate. Be a poll worker. Understand how this works. Ask questions. Ask me questions. Ask all the other experts questions like, we'll gladly answer them, but let's not do this weird conspiracy theory stuff because it actually hurts people.

  Mila Atmos: [00:40:06] Yeah, no, that's that's a really good point. It really... Not only does it hurt people, but I think in the larger picture it degrades faith in democracy, right. Like really at the end of the day. And then it contributes to more voter apathy, when in fact we have just discussed at the top of the hour that every vote counts. And if we were to just do that, you know, two more people in a place like California and some of these really, really tight races, even two more people, ten more people to not vote would make all the difference.

  Ben Adida: [00:40:40] You're making a really important point, which is, you know, democracy is not something that is a natural state of being. And if we don't tend to it, we'll have it forever. No. Democracy is something we have to tend to regularly and feed and care for. And that involves, you know, strengthening our election processes, but also not trying to take them down over rumors. Right. Not trying to reduce trust in the election because it's exactly what you said. It's if you attack this stuff, there's plenty of citizens that don't have the time to dig very deeply into these issues, and they will lose faith in the election process, and they will vote less often. And then we chip away at democracy. And one thing that's very clear from the research is leaders of political parties, leaders of not political parties, but voting organizations. Anybody who's involved

  in that, leaders play a critical role. If leaders go along with conspiracy theories, that has a really bad outcome, it amplifies it significantly. And if leaders say, you know, I hear you, some people think, but no, I don't think that it kind of kills that. It kills those bad rumors. So the call out here is especially to leaders of political parties and leaders of people who have influence. Think about what you say.

  Mila Atmos: [00:41:59] Yes. Well said. So every week on Future Hindsight, I ask my guest to share a civic spark. One small step we can all take to be more empowered and ignite collective change. What's a good way to turn the insights that you've shared with us into action?

  Ben Adida: [00:42:17] I think the easiest way is to talk to your county clerk or your town clerk. If your elections are administered at the town level and say, how can I help? And maybe they need a poll worker, or maybe they need somebody to help with, you know, stuffing ballots and mailing them out or, you know, any of those things. Your clerks are overwhelmed with work and they're having a hard time recruiting people. And if you just spend a day working in the polls, one, you're going to help. Two, you're going to learn a ton. And so that's the recommendation I always give. Be a poll worker for a day. Ask your clerk how you can help. It's not that hard. And it means a lot.

  Mila Atmos: [00:42:55] Yeah, that's good advice. I have to say that I tried to be a poll worker this year, but the only person that they wanted was like a poll worker leader who had to show up at 4:00. And I was like, I'm not doing that. Sorry.

  Ben Adida: [00:43:09] You know, you're right. I should say it is a long day. And if they do ask you to be a poll captain, yes, you're going to have to be there for for a full day. It's true.

  Mila Atmos: [00:43:16] Yeah.
  Ben Adida: [00:43:16] That's true.
  Mila Atmos: [00:43:17] Like, it's a really long day. It's not like. Ben Adida: [00:43:18] It's a very long day.

  Mila Atmos: [00:43:19] you know, 10 to 2 or something like this. It's not like a little shift. It's a whole whole ordeal. So it's a little bit of a depressing time in my mind after this election, 2024. But I really also think that it might be an opportunity for all Americans to seize this moment and make the US a full and vibrant democracy, for example, for more people to vote. So I'm hopeful that Americans will all join in and and participate more fully. But how about you looking into the future? What makes you hopeful?

  Ben Adida: [00:43:55] What makes me hopeful is that, and of course, I'm very focused on, you know, the things that I work on, which is voting, technology and whatnot. But what makes me hopeful is even though there were all these terrible election denial conspiracy theories, voting machines, voting technology is now a topic of conversation, which means people are thinking about it. Which means people care. And there's a lot of work to do to walk back from these. And actual election denialism theories and whatnot. But the fact that people are paying attention, that's probably a good thing for the long term. We do need citizens who are invested and involved in democracy. So that's that's what makes me hopeful.

  Mila Atmos: [00:44:40] Me too. Thank you very much. That is indeed hopeful. I think when people pay attention, like you said, I think they care. And that's a hopeful sign. Thank you very much for joining us on Future Hindsight. It was really a pleasure to have you on the show.

  Ben Adida: [00:44:51] It was wonderful to be here. Thank you so much.
  Mila Atmos: [00:44:53] Thank you. Ben Adida is the executive director of Voting Works,

  the only nonprofit voting system vendor in the United States.

  Remember that civic action doesn't have to be complicated. It's about small steps that spark progress. Like sharing this episode with a friend. Let's recap this week's Civic Spark and fire up our collective power. Ask your county or town clerk how you can help during the next election. Maybe that means being a poll worker. Or maybe that means mailing out ballots for people. You never know what kind of help they need. I promise we're not a broken record here, but we've said it before, and we'll say it again: start local. Your hands on deck are really going to matter there.

  Next week on Future Hindsight, we're joined by Nancy Rosenblum. She's the co-author of Ungoverning: The Attack on the Administrative State and the Politics of Chaos. And she's the Senator Joseph Clark Professor of Ethics and politics and government emerita at Harvard University.

  Nancy Rosenblum: [00:46:04] There are a lot of ways to fail to govern a democracy. Well, there can be obstruction. There can be uncompromising-ness. There can be clientelism, there can be corruption. There can be corruption at scale. We call it kleptocracy. But ungoverning is its own really dramatic, thorough, comprehensive thing. Ungoverning is the intentional destruction of the capacity of government. It's the intentional vandalisation of the machinery of government, and it's unprecedented.

  Mila Atmos: [00:46:37] That's next time on Future Hindsight. Now hit that follow button on your podcast app so our episodes will stay in your rotation every week. This episode was produced by Zack Travis and me. Thanks for tuning in. And until next time, see clearly, act boldly and spark the change you want to see.

  Democracy Group: [00:47:06] This podcast is part of the democracy Group.