
Writing stuff occasionally at mirror.xyz/brunny.eth

Previously Head of Ecosystem team at Aztec Labs

CEO @aztecnetwork. Building a privacy-first L2 on Ethereum. Cryptographer, software engineer, ex-particle physicist. Plonk co-author. Huff inventor.

Thanks again to @zac-aztec and @brunny for coordinating on a AMA! Will do my best to add all of this responses here.

I think they are a great and necessary direction. Plonkish crypto with FRI is already quantum-proof if done correctly.

Constant proof sizes with post-quantum crypto is an extremely hard problem, it might even be impossible. Hopefully we can improve on the state of the art soon.

Working on building zkTLS infra/products. A ruthless empiricist.

http://opacity.network

Advisor @ EV3 Ventures

https://warpcast.com/zac-aztec/0x7fb53fc2

Whelp, we're currently at 32 seconds. But we're doing an optimisation spike this quarter and I'm hopeful we can improve up to 4-5x.

https://warpcast.com/zac-aztec/0x864808eb

Not really.

I think the real value in zk is privacy-focused applications where proofs are generated client-side. It will be a long time before zk chips are found in phones.

For server-side zk, it is a scaling and coordination problem. Compute is cheap. Parallel zk provers are easier to scale than custom hardware.

https://warpcast.com/zac-aztec/0xd3e02763

I think they are an essential piece of infrastructure.

A general purpose privacy rollup is a universal issuance and settlements layer for private assets

App-specific rollups will be how many of these assets will be traded and utilised

They provide much better latency and transaction volume.

https://warpcast.com/zac-aztec/0x7d07a43c

The way I see it; the potential market is so many orders of magnitude greater than the existing market. The goal is to grow the pie, not compete amongst each other for scraps.

https://warpcast.com/zac-aztec/0xa7855b77

I think what you're discussing is similar to a verifiable delay function. They are useful in theory but in practise nobody has produced one that works. They are also very vulnerable to improvements in state of the art compute.

https://warpcast.com/zac-aztec/0xdd88fb60

There are no competitors. There is only Aztec, and those too weak to use it.

https://warpcast.com/zac-aztec/0x7b44c9b1

Aztec is a neutral network so we don't want to play favourites. That being said, we'll happily dedicate resources to any exciting project or application that can accelerate zk.

We also put our money where our mouth is as we are already sponsoring fc hackathons to the tune of $25k!

https://warpcast.com/zac-aztec/0x56bce6cb

I'm very skeptical about quantum computing. It always seems 10 years away.

AFAIK the decoherence problem is monstrously hard. Every extra qbit in your system doubles the difficulty of maintaining stability. We're a long way from using quantum computers to crack RSA keys.

https://warpcast.com/zac-aztec/0x29e7175a

Bottlenecks are memory consumption, the overheads of working with finite fields when proving statements and the cost of making cryptographic commitments.

We already have "linear-time" zk and I am hopeful that eventually we can get the overhead of making a zk proof to be ~100x that of the underlying computation

https://warpcast.com/zac-aztec/0x22166b32

Yes we certainly can play a role!

zk-middleboxes in fc, and used in frames. Proofs that information came from a 3rd party source (e.g. siphoning data out of other social media apps)

privacy-enhanced reputation systems and reputations proofs could be very valuable as well

As for how we can play a role? noir-lang.org

https://warpcast.com/zac-aztec/0xf236cd3d

Get back to me in 9 months! We are on an optimisation sweep with our latest proving system, Honk. I'm hopeful we can get *very* close once we're done.

https://warpcast.com/zac-aztec/0x043f8740

Creating permissionless money and securing human freedom.

https://warpcast.com/zac-aztec/0x016b25af

Private censorship-resistant, programmable money.

https://warpcast.com/zac-aztec/0x2481bd1f

A privacy-first protocol should be modelled on the internet IMO

Fully permissionless and distributed with no nexus of control

We have precedents that the site of regulation for such networks is at the application layer. We want to ensure devs on our network have the tools to distinguish between good and bad actors.

https://warpcast.com/zac-aztec/0x15983791

Well I applied for a bunch of physicist jobs and got rejected from all of them!

But in all seriousness, I also wanted to build tangible things and work in the "real world". That led me to startups which led me to privacy which led me to cryptography.

https://warpcast.com/zac-aztec/0x53dca949

More open and public.

One of the great tensions of our era is the fact that social media has caused a ton of low-information and low-education people  to become politically active. Personally I think this is a good thing and part of civilisation's growing pains.

We are going to be more connected than ever.

https://warpcast.com/zac-aztec/0x5a25dd28