Technological Advancements (and Risks) of Election Processes in the Age of Voter Suppression and viruses
This article is a part of the Tech in Policy publication. TiP focuses on technology being used for good and shines a light on its more malicious or neglectful implementations. To read more, visit this link.
Starting with the Iowa caucus in February, each state primary this election season has exposed more flaws in our current systems than the last. From malfunctioning technology to the pure havoc wreaked by Covid-19, it’s clear that our election processes are not sufficient. Over 80 of Atlanta’s polling locations were closed or consolidated due to fears surrounding the virus, forcing voters to wait in line for hours in the hot sun. Similar scenarios have played out in state after state — some voters in Milwaukee waited in hail.
The pandemic is certainly not the only cause of major election-related failures in the US. Our systems fail us too. Voter suppression affects large groups of eligible voters nationwide. Black voters wait approximately 45% longer to vote than white voters do, Latinx voters 46%. In Florida, minority voters were “more than two times as likely to have their mail-in ballots rejected as white voters.” Our systems are in need of updates across the board.
Barriers that make it difficult to vote or prevent people from voting are part of the problem, but once votes are cast, the technologies used also pose a set of challenges and risks.
Aside from hand counted paper ballots, four types of voting equipment are used in US elections. Ballots can be tabulated using an Optical Scan Paper Ballot System, which, as you may have guessed, scans a paper ballot. Direct Recording Electronic (DRE) systems are devices that record users’ votes directly to the computer memory, sometimes being accompanied by a Voter Verified Paper Audit Trail (VVPAT). Ballot Marking Devices, like DREs, allow users to interact with a UI to make selections, but votes are then transferred to paper ballots to be processed. The fourth type of equipment used in US elections consists of hybrid combinations of Optical Scan Systems, DREs and Ballot Marking Devices. None of these systems have proven to be fully secure, and we saw proof of that in 2016 when our presidential election was tampered with by Russian hackers.
One of the early electronic voting systems was STAR-vote (“Secure, Transparent, Auditable, and Reliable”). STAR-vote was developed from DRE tech and uses end-to-end cryptography where each vote cast creates a 16–20-digit hash string representing the encrypted data that gets recorded in the system. Decryption keys are not kept within the system. Hash values (Zi) are created using the following equation:
where H represents the hash function, E represents encryption, k represents threshold value, v represents the voter’s selections, m represents the machine used, and Zi-1 represents the previous hash created by that machine. STAR-vote is not connected to the internet and it uses a VVPAT which improves reliability, but there are still various points of weakness — machines can be stolen (or break), ballot boxes can be targeted, and hashing algorithms can be cracked. And, of course, voters are still required to show up to the polls in order to utilize an electronic voting system like STAR-vote.
The question is: can the solution to these problems be coded? After all, we already use our smartphones for secure transactions all the time — I just deposited a check on my banking app today. Over 80% of American adults own smartphones, so, why can’t we use them to vote?
Voting by smartphone would solve certain accessibility issues, but it also poses obvious (and not-so-obvious) security risks. David Dill is a Computer Science professor at Stanford and the founder of the Verified Voting Foundation. He’s very outspoken against the idea of mobile voting and claims that with current technologies, there’s no way to guarantee bug-free, unhackable software and hardware. Various electronic voting systems have been tested since 2000, and Dill argues that none have come anywhere close to the security standards that would need to be met for a process as high-risk as the US presidential election.
The decentralization and democratization of blockchain technology has sparked conversation about how it could potentially be used by proponents of electronic and internet voting. Voatz is a platform for mobile voting that makes use of the “security built into the latest versions of smartphone technology, and the immutability of the blockchain” (if you aren’t too familiar with blockchain technology, check out this short video). In order for users to be able to cast their ballots on Voatz, they must be verified by their election jurisdiction and sent a mobile ballot. The mobile ballot contains a set number of “tokens,” each of which represents a single piece of data. On the Voatz app, the user records their selections, altering their given tokens, and submits their final ballot. The authenticity of the submission block must then be verified by multiple distributed validating nodes, each containing a record of the existing blockchain, before it’s added to the chain. Once added, the tokens are debited from the voter’s ledger and added to the correct candidate’s ledger, and all verifier blockchains are updated. Each block on the chain contains a hash value of the previous block’s value, so if any data is changed, the rest of the chain will be affected.
Voatz has already been successfully used in various city and county-wide elections in the US, but there’s a big difference in risk between local elections and the presidential election. In order for us to rely on a mobile voting solution to modern accessibility issues, we would need to be fully confident in the security of our votes. Our smartphones, as advanced and impressive as they are, are susceptible to viruses and malware, which means that they can’t yet provide us with full confidence that votes have not been tampered with. Although mobile voting has its fair share of critics (for good reason, at this point in time), the incredible speed at which modern technology moves means we might not be so far from a trustworthy solution after all.
In the meantime, if you’re interested in improving voter accessibility, here’s some information about how to advocate for voting rights in your state.
