Online voting still faces security issues
- By Mark Pomerleau
- Mar 25, 2015
For those interested in expanding voting access by allowing voters to cast their ballots over the Internet, one government expert/activist has bad news – the security and privacy risks associated with Internet voting won’t be resolved anytime soon.
David Jefferson, computer scientist in the Lawrence Livermore’s Center for Applied Scientific Computing, has studied electronic voting and security for more than 15 years. He believes “security, privacy, reliability, availability and authentication requirements for Internet voting are very different from, and far more demanding than, those required for e-commerce.” In short, voting is more susceptible to attacks, manipulation and vulnerabilities.
Some champions of Internet balloting believe the safeguards that protect online shoppers from hackers can also protect the sensitive information and meet the legal regulations associated with voting online. Advocates further believe that Internet voting will increase turnout, cut costs and improve accuracy.
Jefferson refuted these claims by asserting that there currently is no strong authentication or verification solution for online shopping. Also, while proxy shopping is a common occurrence and is not against the law, proxy voting is not allowed.
Other ideas that have been floated regarding Internet-based voting involve creating an open source platform. Using tools such as the Trust The Vote Project, voters can be registered, ballots can be designed, results can be more easily shared and auditing can be made easier.
Some believe that moving voting systems to an open source platform would not only allow groups to verify system integrity, but assure accountability and voter access. Yet vulnerabilities that can be exploited as a result of open sourcing this type of system are back-door type breaches such as the Heartbleed bug.
Some states have started to experiment with email voting, which Jefferson said is especially vulnerable to attack as email headers can be easily forged, email does not use end-to-end encryption and it does not offer reliable authentication methods. Emails can also be manipulated in transit by IT personnel who control its path. Malware can also be injected into receiving vote servers.
Moreover, fraudulent ballots are extremely difficult to detect because there is currently no reliable system online. Unlike paper or in-person ballots, “Internet elections are essentially impossible to audit, and there’s no meaningful way to recount because there are no original indelible records of the voters’ intent against which to compare the outcome,” Jefferson stated. “The only vote records are on the server, and they are highly processed electronic ballot images that have been operated on by millions of lines of code on the client device, during transit through the Internet and on the server and canvass systems.”
So far, the best models for secure Internet voting are called end-to-end auditable cryptographic protocols, and they are still in the research and development phase. This system works by using advanced cryptographic methods for protection in privacy and prevention of undetected loss of votes, undetected changes in votes, forged votes, miscounting of votes and allowing voters to verify their ballot. Shortfalls and concerns related to these systems include the inability to address remote voter authentication and malware defense.
Electronic poll books can assist in auditing processes. These books enable review and voter information processing, but they do not count votes – they simply assist in monitoring. Thirty jurisdictions have currently used these books.
Mark Pomerleau is a former editorial fellow with GCN and Defense Systems.