 So our next speaker is Amber McReynolds. Watch out. Her talk will be on putting voters first, expanding options to vote. Amber McReynolds is the executive director for the National Vote at Home Institute, and is the former director of elections for the city and county of Denver, Colorado, as one of the country's leading experts on election administration and policy. She's proven that designing pro-voter policies, voter-centric processes, and implementing technical innovations will improve the voting process for all voters. During her time in Denver, the election office was transformed into a national, international award-winning election office. Amber was recognized as a 2018 top public official for the year of the year by Governing Magazine for her transformational work to improve the voting experience in Denver and across Colorado. She's now focused on improving the voting experience across the country. Please welcome Amber McReynolds. Okay. Well, it's great to be here. Again, I was on the panel a couple of hours ago. So great to be here again. I'm gonna talk a little bit about sort of this concept of putting voters first, how that ties into policy, and then hopefully we'll have some time for questions at the very end. But about halfway through, I'm gonna also have another expert come up and talk about auditing, because that's been a topic that we haven't heard a lot about while we've been here at DEF CON so far. So first off, 2018, and these are all pictures that happened in 2018 in 15 to 20 different states. But what you see up here is you see long lines, significant administration problems, machine breakdowns, all of these things. A lot of these, and you can see the paper ballots there, so there's been a lot of talk of this. A lot of the states that had paper ballots still had other problems, long lines because of registration problems, long lines because, say in Missouri, you can only vote on election day. There's no early voting in Missouri, and you can't get an absentee ballot unless you can get a notarized document from your employer or a doctor's note from your healthcare provider. But a lot of these lines, and all of this has been happening, these are the same headlines, same photos that we see year after year in the election process. And I like to show this image because in my mind, a lot of this is because there's basically been games played from a policy perspective because there's been this intensive focus on who wins instead of who votes. So a lot of that has created a lot of these problems. So because voters have not been a priority, and I bring that up in the context of cybersecurity and how we talk about new and emerging voting systems, we have got to put the customer and the voter first in this process because they've never before been first. I promise you, they never have. And because of that, things like HAVA, so HAVA was the Help America Vote Act. That was in response to Florida 2000 and the butterfly ballots. Congress allocated hundreds of millions of dollars to states and said build technology as fast as you can. And that's why we're in the position we are today without dated systems and systems that don't function well. But it's a good example of policy making being reactive, immediately reactive to a problem without actually solving the reasons that voters had failures in their system. Government assigned location. So this outdated notion that we have to assign voters on a specific day to a specific location to vote has been kind of proven to also create issues. So when you think about that, most people now work farther away from where their home is. So if you need to try to get back to where you live and you've got an hour commute, it's gonna be very difficult to get there by seven o'clock. Limited options for voting. Confusing deadlines. Almost every state in the country has different deadlines for voter registration, absentee ballots, mail ballot applications, all of that. It's very confusing to navigate as a voter. Ineffective systems and then sequence problem. And I'll show a slide here in a minute to talk about that sequence problem. So this is kind of the coin phrase I mentioned this earlier, but until we get election policy makers and election administrators and frankly everyone focusing on the who votes and not who wins in this process, we're not gonna see effective change in technology in the process itself. We're not gonna see those lines go down. We're not gonna see any of that. So this is the sequence. And this is what I believe there's been a lot of mistakes made with regards to policy making and that the sequence has been backwards. So basically what we see is people jump to technology and say spend money on the technology. It will solve all the problems, right? And as many of you are technologists in the room, you know that the way the process is and documenting that and developing requirements around that and figuring that out first is really key to making sure that the technology works. It's garbage in, garbage out kind of thing, right? So this is what we did in Denver and then Colorado when we passed our reform. This is a similar model to California has kind of gone the way that Colorado did. And we're seeing a lot more emerging policy discussions that are trying to put the voter front and center and doing pro voter legislation, then designing voter-centric processes and then technology and that's ultimately how we can improve the voting experience. So this, the model I mentioned with Colorado, these are sort of the elements. These are the big key elements that were part of it. We got rid of all the registration deadlines. You can register and vote same day on election day at every election now in Colorado. That'll be the same way for the entire West Coast. So the West Coast will all provide that as an option going into 2020 as well. Ballot delivery, we automatically mail a ballot to every voter before every election and then we offer in-person voting at vote centers. Ballot drop-off options. So what we have also found in Denver is while we deliver the ballot out through the post office right now, most voters about 80% wanna drop that ballot off in person. So we give them 24-hour drop boxes, drive-up drop-offs. You can go to a voting location if you want to and then we have vote centers. Well, instead of having 250, 300 polling locations, we have 30 large vote centers and there's a couple things that have happened with that. We don't need as much equipment. So our equipment needs got reduced significantly. So when we went to replace our voting system, instead of it costing $10 million, it cost $1.5 million. We also have been able to get larger locations like rec centers that are more accessible, that are on bus lines and mainly because we don't need so many spots, we can try to pick the best of the best voting locations. And then effective audits and Jennifer will go over some of the auditing techniques that started and was developed a lot in Colorado and now is also spreading elsewhere. So I wanted to share this. So I, as mentioned, I run the National Vote at Home Institute now and I wanted to share this map because it basically shows what turnout looked like across the country. And you can see on here, a lot of people celebrate 2018 as being a very high turnout election and a lot of interest. You can see there are states that had less than 45% of their voting population participate in the midterm election last year. But then you also see the states that are above 60%. And so the national average was 49.5%. The three vote at home states, so the states that mailed a ballot to every voter last year collectively were averaged about 14 points higher than the rest of the country. But you can also see that the states where voting, there's more pro-voter policies, whether it's options for voters, convenient deadlines, all those things, it's darker. So we know that by making and sort of making that customer experience better, we can improve the process overall. And then I wanted to share this. So this is another trend that's happening and this is something that I think we all have to continue to pay attention to. Since 1992, the blue line is in-person voting on election day. The green line is mail ballot and absentee and then early and in-person is orange. And you can see that early and in-person have kind of been on a steady rise, but there's been more of a rise in mail ballots and the blue line continues to decrease. And so we think in the next few years, we're probably gonna see more than 50% of America will vote before election day with one of the alternative methods. And so as we think about security, as we think about these systems, we have to contemplate where the voters are also moving with what their preferences are. And then this is a report, this is a report put out by non-profit vote, but it speaks to the policies around this topic. And the top 10 turnout states all have consistent policies in place that make it easier to vote. The bottom 10 states are all have confusing deadlines. It's, they don't have much early voting. They don't have much mail ballot. They don't have many options for voters. So this is a very stark difference when you look at the high turnout states versus the low. This report was put out about 2018. It's on non-profit votes website. It's also on our website at vote at home, but I would encourage you to look at it because it really does outline the positive policies that have happened that have impacted turnout. And then this is the current. So I wanted to put this up here. There's basically five different sort of categories of vote by mail and absentee balloting. And if I can get everyone to never use the word absentee again, that'd be great. But vote at home is what we're going for. Ballot delivery is also a good term. But I put this up here to really illustrate the variance by state. So in the state of Texas, if you're over 65, you can get a mail ballot, no problem. If you're under 65, you have to prove with a notarized document that you're gonna be out of the county for 12 hours on election day. Missouri has no early voting and they, you need a notarized document with also two witnesses to get an absentee ballot. So I put this up here because voters in many states, it's really hard to navigate this. And this is the kind of policy discussion that we really need to have to kind of reshape the conversation and put voters first. But it also illustrates the variances in terms of options that are available to voters. And I can tell you, it is much easier. If you're an Oregon voter or Washington or now Utah, and you're automatically gonna receive your ballot in the mail, you think about their experience versus somebody in Missouri that's waiting in line for two hours because they didn't have another option other than election day. Very, very different experiences. So as I always say, change is the biggest barrier. Almost every legislator I talk to about these reforms or about reforming their systems, one of the second things they often say to me is would I have won if this were in place? So change is the biggest barrier because we're dealing with policymakers that have a personal vested interest in the outcome. And that goes back to that who votes, not who wins consideration. But it's possible. And this is the final slide, and I'm gonna have Jennifer come up. These are kind of the six things that I think are really important to have in the conversation. Continuously improve, never be okay. That's kind of how election systems got the way they are as sort of there started to be the stagnation around what was acceptable. Empower voters with options and choice, pro voter policies, implement voter-centric processes, design effective technology across the board, all things, and then validate the results. And the validate the results part is where Jennifer is gonna do her thing with audits. So come on up. So there's nothing like having the top portion of this hacked by sliding somebody else in, but I really appreciate Amber asking me to take a few minutes while we really click. I ran local elections in both Utah and Colorado for about eight years. And a year ago, I had the opportunity to start something called the Election Validation Project with support from the Democracy Fund. And the whole focus of the project was thinking about ways that we can increase trust in our elections by creating standards around audits and testing. And I wanna tell you one quick little anecdote. The thing that I loved best about running local elections is after every single election, I got to do kind of what we're doing here. I could sit down with my team and we could start to pull apart everything. We looked at phone calls, complaints, things that were posted on social media. We pulled data, all sorts of analytics to really pull the election apart and try to figure out what went well, what went wrong, where were the problems, where were the areas for improvement. And when we started to implement risk limiting audits in Colorado, that's exactly how we learned how to do it. By first pulling our processes apart, working through the audit process and coming up with something that's ended up being for that particular sampling method, more or less a national standard. So I just wanna leave you with two quick thoughts. One, we are pushing for a culture of being audit-centric. And that's happening and it's really, really exciting. When I started this a year ago, only a handful of states were thinking about this. Now we have almost 20 states in various places along the spectrum thinking about how to better audit their elections. And so you heard earlier today specifically about risk limiting audits and that's been a big focus of my work for the last year, but I want us to think beyond that. So as you're thinking about the vulnerabilities and you're thinking about the thing areas of concern, also help me think about are there other opportunities throughout the election system where we can stop and pause and think about a way to better test pre-election or audit post-election to validate that the election was done in a manner that was open and transparent and trustworthy and all of the things we are aiming for in conducting good elections. So why do we audit elections? We can detect voting system errors and I've listed hacking and misconfiguration and I'm so glad that there are experts here who are focused on this. My experience is that most of those errors usually happen because of human beings. It provides a way for us to give accountability to our voters. So it gives us something tangible that we can show our voters. It helps us deter fraudulent activity, helps us assure the votes were counted and reported accurately and most importantly, just like I mentioned, it provides feedback for process improvement and that's really where we'll see some significant changes in the way that we run elections and that is why when Amber talks about the model in Colorado and the model that she helped to begin and wrote and crafted and really modeled for all of us, that model because it freed up resources, it freed up time, things ran so smoothly, we had a lot more time to think about process improvement and the way that we were serving our voters. So how do we do that? Two quick things, we need to educate and that's difficult because we're taking something very complex and so there's some guides that I've published, you can see the link up there to the website, thinking about the things that are complex, whether it's a statistical formula or some pretty heavy academic white papers and putting that into something that's digestible for local election officials, for policymakers and others who are interested in helping in this space and understanding how RLA's work. I actually started this a year ago with bullet points about the different sampling methods and I just gave a presentation last week or this week, I don't even know what week it is, to state legislators in Tennessee, this slide with one picture and one word is so far the most effective thing that I've been able to use to communicate how this particular type of sampling method works. So we're still a long ways in taking these complex issues and ideas and bringing them down to their essence and really helping just the lay person understand how it works and it's important that they understand how audits work because if they don't, they're not gonna get behind it and they're not gonna support it both policy-wise with the resources that we need to do that. But the best way and again, I feel like this is a takeaway for me from coming to this conference is to just have people roll up their sleeves and give it a try and so the most effective thing right now that we can do is go around state to state, these are some pictures of some of the audits I've helped with and just walk and talk through the process, very beneficial. Something to think about, RLA's need software we're gonna audit on a statewide level and so that's another vulnerability, right? We're introducing another element of computing and things like that. That's gonna be the case with any type of audit that we do. So I'll leave you with my list. We don't wanna just audit about vote telling system and how ballots were counted, that's super important. That's why I've put most of my work and time and effort over the last year into that but you can see the list up here. There are other things that we need to be thinking about if we want to validate the election as a whole and what I'm finding is that most of those, let's take the top one for instance, photo registration database audits, if we wanna do that, we wanna do that on a statewide level, we're gonna need software to assist us to do that and so engaging with this group is fantastic because having that software open and accessible and able for people to vet and test and sort of validate, I guess, will be really important to this work moving forward. So thanks again, Amber, for giving me the opportunity to stand up here for a minute and evangelize about audits. That one? In ballots though. So most of the anomalies that you see, like if you see a discrepancy is what you're talking about. So meaning the physical ballot doesn't match the way the voting system interpreted that ballot. Most of those are hand marked paper ballots and it's a voter intent issue, right? So you might have had that ballot adjudicated by a team, maybe a marginal mark and they discerned it one way and the voting system may have interpreted, or the audit team may have interpreted a different way. Does that make sense? We still don't have a lot, we don't have a statewide example really of an entire state with ballots that have been cast on a ballot marking device so there's no voter intent issues there, there's nothing, we don't have to worry about the density of the mark. So obviously most of the pilots we've done around those, there have been no discrepancies. The way the voting system interpreted the ballot matches what we see printed on the ballot. So there was one case, I'll give you an example, trying to think of one example where something went wrong is we had like a smudge or a bend in a timing mark or something on that summary ballot. And we weren't able, the system wasn't able to read that. So, any other questions? Wait for her. Anything? Okay, well thank you all for coming to the session and hopefully it was helpful.