Georgia HB 316 is currently working its way through the legislative process. The bill is rather substantial and touches on many different aspects of voting in the state.
I want to start by making a couple of things clear from the outset:
- I based my analysis on security-related concerns as it relates to the voting and vote-auditing process.
- I leave other aspects of the bill, including budget, voter registration management, and political commentary to others who are more qualified than me.
Georgia currently uses direct recording electronic (DRE) voting systems across the state. DRE voting systems are antiquated, and there is near-universal agreement that the state of Georgia should replace the equipment. Additionally, Georgia DRE voting systems do not produce a paper trail. This lack of a paper trail is troubling on two fronts. First, the lack of a paper trail denies voters the ability to verify their vote was accurately recorded. Second, the lack of a paper trail denies election officials the ability to conduct any post-election audit to validate results.
A section of HB 316 seeks to replace the existing DRE equipment with ballot-marking devices (BMD). BMDs have a touchscreen that voters use to select their candidates of choice for the various races. Once the voter makes their selections, the BMD produces a paper record. This record contains a human-readable list of candidate names selected, as well as a computer-readable barcode. It is this barcode that gets scanned to record the official vote.
The problem with this process is that the voter cannot read the barcode, and thus cannot verify that the data contained there is accurate. The voter has to trust the software being used to generate the barcode on the paper record. I am not aware of any voting system vendor that has released any of their software for public review. All that we have to work with is assurances from these vendors that the code is secure and incapable of being attacked, and that they take security seriously.
Voting security experts have raised concerns with the use of BMDs with barcode outputs, and assert they are not trustworthy enough at this point to warrant use in future elections (Alex Halderman, Matt Blaze, and others). In 2018, the National Academy of Sciences released a study on how to best secure the voting process. Their recommendations, which have been echoed by other voting security experts, are to use hand-marked optical-scan paper ballots and risk-limiting audits.
The current bill under consideration by the Georgia Assembly is a step in the wrong direction, in my opinion. Moving the state towards BMD usage with barcode output goes directly against the recommendations of multiple experts on the topic, and does little to improve the security of our voting process.
I hope that the Georgia Senate will fail to pass this bill as currently written and that our elected officials will instead listen to the collective input of experts on the matter.