3 ms·
The bigger the county, the tougher the job. That's why I'm saying this is so difficult; for a realistic hack you'd want to change as few votes as possible. If y
by burkaman 8y ago
The bigger the county, the tougher the job. That's why I'm saying this is so difficult; for a realistic hack you'd want to change as few votes as possible. If you really could predict the whole election in advance, you could pick and choose small counties with defeatable systems.
Targeting big cities solves the prediction problem, but makes your hack infeasible. Phoenix is in Maricopa County, which used two different types of voting machines in 2016. One of them is a paper ballot optical scanner, which prints out a paper tally at the end of the day. So you'd need to hack hundreds of these airgapped machines, and predict the results well enough to not trigger a recount with your changes, since there are paper ballots to compare against. I think this is just as impossible as choosing key counties in advance.
- everdev 8y agoI guess I'm not familiar with how votes are tallied. Are results visible to election officials in real time as people cast ballots or only at the end? Wouldn't it be feasible to look at a midterm election or registered voters (say you had a 70/30 D/R split) and pre-set a machine to return a 60/40 result?
- burkaman 8y agoDepends on the system, but Maricopa County used this machine in 2016, where a polling place official only sees a tally at the end: https://www.verifiedvoting.org/resources/voting-equipment/sequoia/optech-insight/ https://www.verifiedvoting.org/resources/voting-equipment/se... Your idea is feasible if you can break into a warehouse and hack hundreds of these machines, but do you think it's possible to change ~150,000 votes, resulting in a a 10 point swing from recent polls and exit polls, and remain undetected? Remember you're in a big city with lots of news organizations and big elections departments.