Don't Let Iowa Bring Our Elections Back to the Stone Age

We can all agree that the Iowa caucuses were a mess. So much so that some people are calling for the removal of all election technology. Information is still trickling in about what went wrong with an app that was designed to report in-person caucus results to party leaders. Some caucus participants said they couldn't download the app, while others didn't know how to use it. What we do know is that this app was developed in secret, over a short period of time, and without the benefit of review by government agencies or independent security teams.

Basically, Iowa did everything wrong when it comes to deploying a new technology — especially a new technology being used in such a high-stakes election.

Having worked many years as an FBI Special Agent on cyber threats from nation-states, and as a security expert who has reviewed dozens of mobile voting systems for both their technical and security capabilities, it's easy for me to see why people are comparing Iowa's app to mobile voting and calling for paper ballots. But we shouldn't. In fact, eight mobile voting pilots in six separate jurisdictions around the US have been conducted safely and securely with audits, no security breaches, and a paper trail (locations include Seattle, Utah County, and Denver).

First and foremost, a real mobile voting system actually involves voters using the system to cast votes. A real mobile voting system has an established two-step process for authorization and authentication between the voter and the jurisdiction. It provides a pseudo-anonymous experience, in which the voter marks a ballot and securely submits it to their jurisdiction, where it is received, acknowledged, recorded, and later tallied among the many other ballots. More importantly, a real mobile voting system will have been tested, again and again, with constant input from local election administrators who are acutely aware of how technology can and must work for their voters.

At the core of mobile voting is its ability to take advantage of the confluence of decades' worth of technology advancements and best practices to create a secure environment for the voter. From biometric authentication and end-to-end encryption to cyberattack mitigation and third-party penetration testing, security best practices have been baked into smartphones used by millions to safely make purchases, conduct business, access bank information, and secure their homes. And while not all phones and applications are created equal, it is the ability of engineers to access the security features provided by system manufacturers that allows for these technologies — when leveraged properly — to resist attacks, detect intrusions, and protect sensitive data. 

This is not to say that mobile voting systems are immune to attack more than traditional election systems. We all know about voting machines breaking down, vulnerabilities in voting kiosks, unencrypted fax and email ballots by military and overseas citizens, and the infamous hanging chads and butterfly ballots in the 2000 general election. But to compare all election technologies to an app that was scraped together in two months and failed to undergo testing or training belittles the hard work of election vendors across the country for painstakingly building voting technologies that undergo rigorous auditing and testing.

The lesson from Iowa should not be for election officials to turn their backs on technology. Our takeaway should be that new technologies need to undergo strict testing by government agencies like the Department of Homeland Security, independent security firms, and the white-hat hacker community at large to find vulnerabilities in advance so they don't happen on election night. We should start small so we can test the concept and strengthen security capabilities in controlled settings. Security is an iterative process that gets better over time. There is no room for error in our elections, especially when it comes to data leakage, compromised encryption, broken authentication, or denial-of-service attacks.

I am from Colorado, a heavy mail-in ballot state. My father prefers to vote in person. My mother exclusively votes by mail. My brother uses his smartphone for pretty much everything. Mobile voting should match the voting experience of those who vote in person, by mail, or over the Internet, and must always be just one of many voting channels available to a voter. Let's not allow one bad attempt stop us from finding new ways to achieve this.

Related Content:

Check out The Edge, Dark Reading's new section for features, threat data, and in-depth perspectives. Today's featured story: "8 Things Users Do That Make Security Pros Miserable."