Last Wednesday, after a midterm election that attracted the highest turnout for any midterm in more than 50 years, defeated independent candidate Tim Canova posted a video to his Twitter account. The footage, captured by a concerned citizen, appears to show election workers arriving in their personal vehicles and placing provisional paper ballots into a rented box truck. Each person seen delivering ballots reportedly arrived alone, without the required second person to witness the transfer as mandated by chain-of-custody rules.
The video is striking. It was recorded after polls had closed and in darkness, giving the impression that something irregular was occurring. The citizen who filmed the scene can be heard saying, “I don’t think that’s right,” a reaction many viewers likely shared.
For political observers, the clip served as a stark reminder of how delicate public trust in paper ballots can be. But the location where the video was shot surprised no one paying attention.
Caught On Video: Concerned citizen sees ballots being transported in private vehicles & transferred to rented truck on Election night. This violates all chain of custody requirements for paper ballots. Were the ballots destroyed & replaced by set of fake ballots? Investigate now! pic.twitter.com/GUPZnz7Gny
— Tim Canova (@Tim_Canova) November 8, 2018
The footage originated in Broward County, where Canova lost by a wide margin to Debbie Wasserman Schultz. Broward is also the county overseen by Brenda Snipes, who has a history of controversies related to election administration.
When Canova first challenged Wasserman Schultz in the 2016 primary, Schultz had already gained national notoriety after stepping down as DNC chair amid revelations that the Clinton campaign exerted financial influence within the party. After Canova lost that primary, he exercised his legal right to inspect the ballots. Snipes attempted to charge his campaign tens of thousands of dollars for access to those ballots, which led Canova to file suit.
Subsequently, Snipes authorized destruction of those ballots despite a federal court order instructing her not to do so. She signed paperwork indicating there were no pending lawsuits related to those ballots and authorized their destruction a year earlier than standard retention rules required. She later called the action a mistake.
A federal judge found the destruction illegal and said that Snipes’ dismissal would be appropriate, yet Governor Rick Scott declined to remove her from office.
Since Snipes took office in 2003, Broward County has repeatedly been among the last jurisdictions in the nation to finish counting votes, despite having a head start over West Coast states.
Given that track record, the video showing questionable ballot handling in Broward County under Snipes’ supervision surprised few observers. After the highly contested Senate race concluded, Governor Rick Scott—soon to be a lame-duck governor—filed suit against Snipes alleging possible election irregularities. President Trump also weighed in on social media, calling the situation an embarrassment and using the phrase “election theft.” A judge later ordered Snipes to allow immediate viewing and copying of the ballots in question.
While I do not typically align with the President, the combination of this video and the histories of Schultz and Snipes makes it difficult to ignore the appearance of impropriety.
Other states, including Georgia, faced separate accusations related to voting irregularities, long lines, and concerns about the security of voting machines.
Blockchain technology offers a potential solution to many of these problems. Implementing a voter-verifiable, blockchain-supported system need not be complicated for voters. The basic idea is to combine transparent, open-source voting machines with a public blockchain anyone can use to verify that votes were recorded.
Imagine arriving at a polling place where officials check your name against the voter rolls and hand you a ballot as usual. After filling out the ballot, the voting terminal asks you to create a password of at least eight characters. That password, combined with your voter roll information, generates a unique cryptographic key or “voter number.” You then insert your ballot into the machine; the system uploads a hashed representation of your vote to a public voter blockchain and returns the associated voter number. Later, you can visit a government-run block explorer—or a third-party alternative—and use that voter number to confirm your vote appears in the ledger. No one could determine how you voted unless you disclose the voter number that ties you to that entry, and you could prove your connection to that vote with your personal credentials if needed.
In time, voters might be able to cast ballots remotely from their personal devices. That would require voters to manage private keys themselves, however, which adds complexity and security concerns. For now, a hybrid approach—generating key pairs at the polling location using voter roll data plus a simple password—would provide an extra layer of verifiability without placing the burden of key management on every voter.
No system is completely immune to manipulation. An official could, in theory, generate voter numbers for absent voters. But existing oversight measures that prevent physical ballot stuffing would still apply, and it is generally difficult for a single worker to submit multiple ballots unnoticed. If necessary, keys could instead be generated at registration to further reduce that risk, though that would require voters to retain or store credentials securely.
With a properly implemented blockchain-based system, destroying votes would become essentially impossible. Recounts would largely become redundant, since every vote would be publicly recorded and verifiable on the ledger. Anyone could audit the results manually or run their own scripts to tally votes.
To be a functioning blockchain, the system requires participants who secure the network. In cryptocurrencies, miners are motivated by financial rewards, but in a voting blockchain the primary incentive would be protecting the integrity of elections. Political organizations, civic groups, researchers, and individual citizens could run validator nodes to support the network. If necessary, additional incentives—such as tax credits for running verification nodes or using established blockchains that already have security and incentives—could be explored, though tying voting infrastructure to third-party public blockchains may introduce transaction costs.
The concept of blockchain voting is not new; decentralized communities and digital platforms have used blockchain-based voting mechanisms for governance for years. Proposals to put public elections on a blockchain have surfaced before, and some projects are already exploring decentralized digital voter IDs and on-chain vote storage. One such effort, Elph on Ethereum, aims to provide digital voter IDs and vote storage, though I share concerns about the risks of lost or stolen digital credentials.
I lack the technical skills and connections to build this myself, but I’ve discussed the idea with industry professionals. Nathan Wosnack, founder and CEO of Ubitquity, which has experience recording documents to blockchains, told me the idea is feasible and not particularly difficult to develop. He noted that his company already uploads hashes of records to blockchains for clients and could apply similar methods to voter records.
He assured me he wouldn’t appropriate the idea, though I actually hope someone takes it up. A working implementation could reduce the threat of election manipulation, eliminate contentious recounts, and raise public awareness of blockchain technology. If blockchain can secure our elections, it could also help popularize broader adoption for other applications, including financial services.
I can’t think of a better outcome for the country or for the cryptocurrency community than a secure, transparent voting system that restores confidence and encourages civic participation.