Nathaniel Fruchter

Trust is built through evidence: elections produce evidence (paper trail, audits, tabulation, certification of processes and devices, outcomes). How much evidence do you need to build trust? #enigma2020

Goal: everyone who votes should be able to trust the outcome of an election. But we don't live in a perfect world! It's an open question of how to achieve universal trust. #enigma2020

For every technical solution, there exists a political side that is a difficulty multiplier. Talk focuses on the technical side though. [NB: Not just a political side, a socio-political side. Politics is also behavior and culture!] #enigma2020

Now an overview of ballot secrecy. Publishing election data is trustworthy, but isn't tenable for many reasons. First potential sol'n: risk limiting audits. How much of the paper trail do you need to examine to build confidence in the reporting? #enigma2020

Another: end to end verifiable elections. We need certified totals and the ability for someone to make sure their vote was counted correct. Instead of revealing each vote, split into buckets with homomorphically encrypted 0s/1s for each person + 0-knowledge proofs #enigma2020

This is hard to explain though: how do we build trust through something that can't be explained easily? Argument about transitive trust: do we have enough trust in experts for us to say "our experts checked the results". #enigma2020

Help America Vote Act pushed new voting machines, but didn't provide the best incentives ($$) for building good critical systems. #enigma2020