The disruptive promise of internet voting

The fact that the agency is putting $10 million into the development of an unhackable voting system signifies some very real concerns over the risk of outside interference in the democratic process.

We’re more familiar with fears over ‘fake news’ and ‘dark money’ – but if someone really wanted to undermine trust in democracy, then they couldn’t do better (or, rather, worse) than to compromise the integrity of voting itself. Once that goes, everything else is called into question.

Kim Zetter reports on the DARPA funded project for Motherboard:

“The first-of-its-kind system will be designed by an Oregon-based firm called Galois, a longtime government contractor with experience in designing secure and verifiable systems. The system will use fully open source voting software, instead of the closed, proprietary software currently used in the vast majority of voting machines, which no one outside of voting machine testing labs can examine. More importantly, it will be built on secure open source hardware, made from special secure designs and techniques developed over the last year as part of a special program at DARPA.”

In contrast to the pen-and-paper voting systems used in the UK, Americans use machines to cast, count and verify votes. That obviously speeds-up the whole process, but leaves it open to the risk of hacking and other forms of hi-tech fraud.

Then there’s the opacity of the system. In Britain, you can see ballot papers being counted by hand – the piles for each candidate accumulating on tables in a room full of observers from the different political parties. Rigging a system like that requires physical actions that leave physical evidence. But when the count takes place on a microchip, any fraud is invisible to the casual observer – and, just as important, so is the absence of fraud. This provides a context in which malicious actors can spread rumours about hidden algorithms and undetectable hacking. Unable to verify claims and counter-claims with their own eyes, the only thing the public can do is decide whose word to trust – not an ideal state of affairs in such a politically polarised nation.

Making the software that runs the voting machines open source – and therefore verifiable by as many different stakeholders as care to look into the matter – would, therefore, seem like a wise move.

It would help if the voting tech was generally reckoned to be tamper-proof in the first place. To this end, the DARPA project is focusing as much on the hardware (i.e. the voting and vote-counting machines) as it is on the software:

“…most hardware is gullible and has no way of distinguishing between acceptable and unacceptable behavior. If an attacker’s exploit tells the machine to do something malicious, the hardware complies without making judgments about whether it should do this.”

The example given is of someone “slipping a malicious memory card” into a voting machine so that it recorded one vote multiple times – “as researchers have shown could be done with some voting systems”.

But while necessary, the emphasis on hardware is also disappointing because the real breakthrough would be a voting app so secure that it could allow elections to be conducted through one of DARPA’s earlier projects – the internet. We can access our bank accounts and make major purchases using our smartphones and laptops, so surely it should be possible to cast our votes in the same way.

In a few countries it already is. Estonia has had online voting since 2005 – and despite the constant threat to the country’s cyber security, the system appears to be robust. In large part, that’s because the Baltic republic has a sophisticated ID card system through which its citizens can easily prove who they are and access a wide range of public services. One could argue that a similar system would be a more productive investment in America’s security than building Trump’s wall.

What would universal online voting do to democracy itself? For one thing, we could afford more of it (once the tech was up and running). For instance, it would be easier and cheaper to run recall elections to remove elected politicians who have let their constituents down. In the UK, the member of Parliament for Peterborough, Fiona Onasanya, was jailed this year after being found guilty of perverting the course of justice. Refusing to resign her seat, a recall procedure has been triggered: a by-election will be called if more than 10% of eligible voters in Peterborough sign a petition. Whether successful or not, the recall petition will cost around £500,000 to administer – with even more expense if there’s a by-election.

And that’s the cost of democracy in just one of 650 constituencies. The direct public cost of Theresa May’s snap election in 2017 was £140 million. Running the Brexit referendum cost about the same. In short, voting is expensive – and thus a practical constraint on direct democracy.

However, the reluctance of politicians to let us vote more often isn’t just about saving money. The argument that “if voting changed anything they’d abolish it” is not only cynical, it’s plainly wrong. Voting can and does change things, which is why the establishment holds as few public votes as it can get away with (and even tries to overturn results when we vote the ‘wrong’ way).

Internet voting would remove the cost barrier to direct democracy, which is the real reason why most governments aren’t rushing to make it happen.

Click here to read our series of answers to the question: how can we fix our democracy?

---

Peter Franklin is Associate Editor of UnHerd. He was previously a policy advisor and speechwriter on environmental and social issues.

peterfranklin_