sugo

Excited to share Bolt, a new multilinear polynomial commitment scheme (MLPCS) with @kobigurk and @ronrothblum ia.cr/2026/310. To appear at Crypto 2026.

@sinu_eth @PrivacyEthereum tlsnotary.org/why I'm guessing it's for this use-case? verifiers are third-party apps, and provers are end-users, who are registered in another service the third-party app needs access to

@sug0x0 @PrivacyEthereum VOLEith has higher prover cost and larger proofs because it can't use LPN-based VOLE generation. SpeakUp is targeting the designated verifier setting, and interaction is not an issue

SpeakUp: a new zkVM we're prototyping for private, client-side proving of WebAssembly programs - built to run on low power devices such as mobile phones and browsers. The underlying proof system explore different tradeoffs for different use cases from mainstream zkVMs. SpeakUp is interactive, and uses vector oblivious linear evaluation (VOLE), which provides a fast linear-time prover and post-quantum security at the cost of larger proofs. See the early design draft below, we're looking for public feedback 👇 pse.dev/blog/introduci…

Introducing Vizor. A beautifully crafted @Zcash wallet for your desktop. Multi-account in one wallet. Keystone on day one. Open source. Available now on macOS: vizor.cash

Looking to learn about lattice-based cryptography? Check out my tutorial and survey: web.eecs.umich.edu/~cpeikert/pubs…

Ran on a Macbook Air M4. Optimizations pending! Paper: eprint.iacr.org/2026/855 Repo: github.com/NethermindEth/… The main ideas are: 👇

@carover0 @vvvdft88435 @jswihart even weights on byzantine consensus would work, just as in literature (everyone has the same voting power), but even that is arguably centralized, and doesn't scale too well. nakamoto consensus scales even worse with many miners with even power

@carover0 @vvvdft88435 @jswihart achieving consensus under a decentralized system is fundamentally kind of a paradox (centralization wise). if all nodes go off on their own, the system is useless (no one agrees on anything). eventual consistency could be nice, but double spending seems to require a total order

A significant number of Zcash miners sell what they receive each day to cover costs and generate profit. At $380 per $ZEC, that’s under $550k. Peanuts. Most Zcash holders are Zodling, and many accumulating as the asset is massively under priced. Not investment advice.

@lunardragon420 @enjojoyy basically just use codex 5.3. I want to try glm/kimi on ppq

@lunardragon420 @enjojoyy I'm using opencode with ppq.ai and copilot. the former has decent api prices, and more importantly, no kyc/phone number bs to register. the latter I get basically for free from my org. when I hit the limits to copilot, I start using the ppq models

I just cancelled my Claude Code subscription to switch to Codex

CHOP DOWN THE DARK FOREST Excited to reveal @commonwarexyz's new Batched Threshold Encryption (BTE) scheme that achieves [1] SOTA concrete decryption complexity (fastest known) and [2] minimal ciphertext size (smallest known).

Don't be surprised when a LayerZero multisig exploit ends up funding some more DPRK tests 🫡 612 days ago

@inner_concerns @lukevalenta Some updates here: #the-zoo-of-post-quantum-signatures" target="_blank" rel="nofollow noopener">blog.cloudflare.com/pq-2025/#the-z…

@paulmillr how many of the issues were introduced by llms? :p

noble crypto v2.2 is out after a full LLM-assisted self-audit. Found more issues than all previous "independent" audits combined - while finding nothing major. AI gives attackers & researchers superpowers. Fast security feedback loops are now critical. Also: FROST & Big Endian.

@zkDragon clients in mastercard/visa are stateless. depending on the layer where privacy is applied, clients require managing state, which is always a bottleneck, not only for throughput but also ux. you can use managed wallets, but give away privacy for convenience

Private payments winning means capacity must literally scale 10000x. Mastercard + visa is 13k average tps, 45k capacity. Today the best system is 2.92 tps throttled by shielded sync. That limiter is getting removed this year. Zcash

Added ML-DSA-44 signature support to github.com/sug0/git-signi… via git.sr.ht/~sugo/ml-signi… git-signify is a project I started a while ago, because I thought GPG sucked as a signing mechanism in git.

ZK for hash-based cryptography is getting solved quickly! This sets the path for privacy applications: fast provers with minimal security assumptions running on mobile devices and web.

Sean Bowe (@ebfull) and Dev (@zkDragon) recently did a whiteboard session to discuss Zcash's post-quantum roadmap: recoverability very soon, using PIR w/ the Tachyon upgrade to achieve full PQ privacy, and setting the stage for PQ soundness alongside major scale improvements.

Intel SGX has fallen! Its most important key is in our hands: we extracted the Global Wrapping Key from an instance of the Intel Gemini Lake platform