Well before the 2000 presidential election, government, academics and industry were testing the Internet as a medium for casting and recording votes. The Florida debacle, with its butterfly ballots, contested chads and equipment failures, gave new impetus to this research. But despite the clamor for more reliable voting systems that arose from the bitter Bush vs. Gore contest, experts say security and infrastructure problems make Internet elections a distant prospect, at best.
After several years debating minimum requirements for voting equipment, the computer science and public policy communities appear to agree that the Internet-as it exists today-can’t sufficiently safeguard the privacy, security and reliability of the voting process. Pitfalls range from the obvious, such as malicious hackers, to the obscure. For example: Every state requires that votes be cast in secret, but how can officials verify that a party hack isn’t standing beside a remote voter? Might Internet voting spell the return of high-tech Tammany Halls? And won’t the system help the techno-savvy more than others, increasing the voting gap between rich and poor?
“There’s just a huge number of problems,” says David Jefferson, former chairman of the technical committee of the California Internet Voting Task Force and senior member of the technical staff at Compaq Computer Corp.’s Systems Research Center in Palo Alto, CA.
Internet voting is already here. At least two companies, VoteHere Inc. of Bellevue, WA, and election.com of Garden City, NY, regularly manage student polls, association elections and corporate proxy votes. They’ve also served limited public elections. In 2000, for example, VoteHere handled 35 Internet votes for Alaska’s Republican presidential straw poll, and election.com processed nearly 40,000 Internet votes for Arizona’s democratic presidential preference primary. And a separate government initiative let approximately 80 overseas military personnel cast their November presidential ballots over the Internet.
“I haven’t heard of any reported security problems with any of these,” says Jefferson, “but that doesn’t mean the election wasn’t undermined” in a way that escaped detection.
Jefferson and other systems theorists point to a raft of problems that arise when long-standing procedures designed to prevent vote-buying and other fraud are translated to the Internet. In traditional polling places, voters can prove their identity by showing, say, a driver’s license. Over the Internet, however, identity is notoriously easy to fake. Proposed solutions include giving voters a unique ID when they register in person, or requiring Internet voters to submit offline paperwork similar to absentee balloting. The California task force concludes that only with such safeguards could Internet voting be as secure as the today’s absentee voting system.
Of course, confirming one’s identity online is not an insurmountable problem. Just look at all the e-commerce sites mushrooming across the Web thanks to electronic credit card transactions. But voting poses a unique problem: elections require that the voter’s identity, once verified, be stripped away to ensure anonymity. Another privacy problem: no matter how secure the Internet voting servers, voters’ own computers would still be vulnerable to hackers. And although those who vote from behind corporate firewalls would be safer from hackers, they would face the risk of network administrators spying on them.
Electronics to the Rescue
But high-tech voting can still improve elections, say researchers, who recommend wider use of two other technologies: optical-scanning systems, in which the voter marks paper that is then recorded electronically, and direct-recording electronic devices-which often use touch screens-that also record and count votes. The technologies address two failures exposed in Florida: first, they allow voters more chances to verify or change their votes, and second, they provide better audit trails.
The Caltech-MIT Voting Technology Project, a joint research effort, proposed last fall that electronic ballot systems record each vote on a physical object such as a cheap memory card. The object would contain more information than a simple ballot; it would also record information about the polling station staff, the precinct and the form of the ballot. The project concluded that separating the design of the object from design of the voting mechanism would encourage innovation in voting machines, create clearer audit trails and make it easier to design user-friendly ballots for different voters, including the blind and non-English speakers.
The Caltech-MIT project estimates that replacing old equipment with optical scanners or with comparable direct-recording electronic devices, and making registration data widely available, could cut the number of lost votes in half by 2004. Projected cost nationwide: $1 billion for scanners, $2.6 billion for touch-screen systems. But cost and political resistance could slow even these more proven technologies. “Localities just don’t have that money and expertise,” says project member Charles Stewart, an MIT political science professor.
The Politics of Science
Federal Election Commission chairman David Mason says those kinds of systems, unlike Internet voting, are likely to meet the standards set by the Commission, which recently released a draft of updates to its 1990 voluntary standards (the new standards should be approved later this year).
“[The guidelines] are designed for computerized technology,” Mason says, adding that engineers will play an important role in shaping the standards. A committee of volunteer engineers is developing uniform performance standards for all types of voting equipment for the Institute of Electrical and Electronics Engineers, says committee chairman Stephen Berger.
Mason, however, doubts that any Internet-based solution can meet the requirements. “It didn’t appear after the California commission and our own investigation that there was anything out there anywhere near ready.”
Even though the likelihood is low that we’ll be voting online, the Internet could soon play important ancillary roles in elections. Researchers say that the Net could provide a low-cost means to transmit registration databases and vote totals; the technology is also well-suited for overseas military voting and other areas where good security is already in place.
But, concludes Stewart, “my hunch is that even when the security issues get solved, Internet voting is going to be a niche.”
A quick guide to the most important AI law you’ve never heard of
The European Union is planning new legislation aimed at curbing the worst harms associated with artificial intelligence.
It will soon be easy for self-driving cars to hide in plain sight. We shouldn’t let them.
If they ever hit our roads for real, other drivers need to know exactly what they are.
Crypto is weathering a bitter storm. Some still hold on for dear life.
When a cryptocurrency’s value is theoretical, what happens if people quit believing?
Artificial intelligence is creating a new colonial world order
An MIT Technology Review series investigates how AI is enriching a powerful few by dispossessing communities that have been dispossessed before.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.