I got this email from a former colleague at UW to a mailing group recommending all of us to use AI.
This is from a remarkable computer scientist, who invented the fastest algorithm for the traveling salesman problem.
OpenAI is accelerating and transforming how science is done with GPT 5.5 Pro.
Excited to unveil our work on this soon.
Our work on sigma-rs and credentials was accepted to USENIX 2026!
Thanks to the @torproject and @OpenObservatory for the real-world problems, feedback, and inspiration that shaped it.
With pq-privacy is solved, the next step is post-quantum soundness.
https://t.co/MwxcKQrJAQ
@veorq that's such an old story, security engineers claiming that proofs and formalism don't matter, and formal people being all snobby in their ivory tower.
and yet, maybe there's a reason cryptography has been doing OK since circa 1976
@yarnf40580 @tqbf what are you talking about, it's not even possible to have a trait implement distinct features for generic groups and fields, or const generics that do basic arithmetic like subtraction. the type checker runs in exponential time.
We found the same Fiat-Shamir bug in six independent zkVMs.
The result: an attacker can bypass the cryptography entirely and prove mathematically impossible statements (like minting $1M out of thin air).
Full breakdown ↓
one day I hope to be a great advisor like Karthikeyan and lucky enough to have students as smart as Nadim who continue working on my research line even after their PhD
MASSIVE rewrite of my Verification Theatre paper now on ePrint with added:
- Three new bugs that are *inside* libcrux's verified core,
- Comparison to Amazon's verified LibCrypto components,
- Many additional sections, discussion, details!
Go read it! https://t.co/yYD6cxSDf0
Former BlackHat board member Vincenzo Iozzo, and co-author of iOS Hacker's Handbook, had a relationship with Jeffrey Epstein.
It appears Epstein attended DEFCON and/or BlackHat in 2013 and 2015, possibly 2016.
The zk agenda is moving!
Just spoke at Linux Foundation's Decentralized Trust, great to see people realizing the lack of specifications in the space and working towards fixing this.
Introducing a new standards lab for generic zk-SNARKs, in collaboration with Linux Foundation Decentralized Trust (@lfdecentralized).
Join us on 14 Jan for our first virtual meetup, along with Mary Maller, @srinathtv, @mmaker, and myself:
https://t.co/TeboLCZeek
My talk at Collège de France is out!
It’s been fun to give a lecture about modern tools for anonymity starting from simple zk tools, and bridge theory of cryptography with real-world authentication systems.
https://t.co/UxTVyKqezf
Presenting my paper on keyed-verification anonymous credentials at @acm_ccs in Taipei in just a few hours! 🗞️ https://t.co/wEVZ9VdSN2
Exciting to present it in the very same venue where I wrote a big chunk of it while attending @rightscon!
@alpeh_v@levs57 p3 is zk and has no_std support (I was aabre to generate a wasm proof). agree that proof sizes are bad :( Curious to know if you end up finding something better, I'm on a similar boat as you