✳flâneur — a map of the web's best reading

What is going on with witness encryption?

blog.aayushg.com · 1,888 words · saved by 1 readers

Thanks to Sora Suegami, Vivek Bhupatiraju, Gavin Uberti, Yi Sun, Jonathan Wang, Flynn, and Florent for thoughts on witness encryption. Updated as of 11/9/2023. Witness encryption is a pretty underrated idea in cryptography that hasn’t been extensively explored or applied (especially in blockchain) yet. Here are some “gradients” of witness encryption: However, it’s important to note that all witness encryption algorithms over NP, including the one implemented by Guberti (and the paper it’s based off of), are susceptible to the zeroing attack. This vulnerability is the reason why multilinear maps do not exist. As of mid-2023, there is no known algorithm for witness encryption that is cryptographically secure. Several papers have proposed such algorithms, but each introduces a novel and unproven mathematical assumption. One exception is the Witness Encryption (WE) from general indistinguishability obfuscation (iO), as discussed in a 2020 paper. However, this approach is highly inefficient

Nov 24, 2023 crypto Thanks to Sora Suegami, Vivek Bhupatiraju, Gavin Uberti, Yi Sun, Jonathan Wang, Flynn, and Florent for thoughts on witness encryption. Originally written in Q4 2022. Witness encryption is a pretty underrated idea in cryptography that hasn't been extensively explored or applied (especially in blockchain contexts) yet. Witness encryption is the idea that you can encrypt not to a person or public/private key pair, but to anyone with any witness (aka satisfying statement) that satisfies some constraint. This can be people with a certain signature, or possibly even a certain ZK

Explore this link on the map →

related reading