Member-only story

Fully Encrypted Computing

David (drbh) Holtz
8 min readApr 22, 2019

Doing computations on data, without ever seeing the input or output. What⁉️

This is my first article since working on the Coinlist Hackathon, and I’m proud to announce that I placed 3rd 🥉 in the tooling section check out the announcement article here, wooo wooo!

Since that hackathon i’ve been crawling around NuCypher’s other GitHub repos and getting my head wrapped around Fully Homomorphic Computing. NuCypher is working on a GPU powered Fully Homomorphic Encryption library that makes FHE finally practical (in the past it took too long to be useful).

Homomorphic encryption is one of the holy grails of computer security. It allows a computer to perform calculations without any leakage of the actual data.

If implemented, it would allow for a new level of security. One where business couldn’t see any of your data, but still add the value you want.

Imagine Instagram could apply filters without ever seeing the photo. Alexa could answer you without possibly leaking the recording to a curious developer. Imagine credit cards could be suggested to you based on your bank accounts without ever seeing the balance ⁉️

How does this trickery work?

Well I won’t dive too deep into the topic, but essentially if you encrypt data in a very specific…

--

--

David (drbh) Holtz
David (drbh) Holtz

Written by David (drbh) Holtz

Prompt/Software Engineer | 🦺 always building 🪛

No responses yet