📣 curvesma.sh is back open for submissions We paused submissions yesterday during the @GitHub outages. Verification is now updated, with a new guide for checking your submission locally. Run yukon sync to get the latest, then follow the updated participation guide before you submit. Thanks for your patience.

very excited aboout the next few weeks! we're pulling no-bullshit privacy tech into the real world

When people ask me why I keep building with @aztecnetwork and @NoirLang after 2 years, I tell them: privacy is one of those things where the difference between "real" and "marketing" is invisible to most users but brutally obvious to developers. A quick background on me: I spent years at Disney + Hotstar building systems used by hundreds of millions of people, and I've now been building in crypto for the last 6 years. I've seen what it takes to build infrastructure that actually works at scale and what it takes to pretend you do. Here's the uncomfortable truth: most "privacy-focused" chains today(you know which I’m talking about) have superficial privacy at best. A centralized sequencer. No client-side proving. A privacy toggle that mostly exists in the marketing deck. That's fine for fooling retail traders chasing the hot narrative, but ecosystem developers see right through it. Real privacy tech is genuinely hard. It requires: • Decentralization from day one, not as a roadmap promise • Client-side proving as a first-class primitive, not an afterthought • Years of unglamorous engineering, audits, and iteration before anything ships That's exactly what Aztec has been quietly building for years. No hype cycles, no narrative-chasing, just deep, foundational work on true programmable privacy. As a developer who's actually built on the stack, I have tremendous respect for the team. One wish: I'd love to see more open, honest discussion in this space about the difference between actual privacy and superficial privacy. The bar for calling yourself "privacy-focused" should be higher. And a special shoutout tremendous respect for @zcash as well. The OGs who proved private digital money was possible. The space stands on their shoulders. Privacy isn't a feature. It's an architecture. Build accordingly.

RT @sreeramkannan: Can AI break ecdsa? Not quantum but right NOW? CurveSmash is live on Yukon. The best way to prepare: be proactive a…

Doing due diligence on every @aztecnetwork sequencer takes time, but you can use AI 8 checks, 4 websites, 1 block explorer… per provider! You could do it by hand, or have your AI agent do the legwork with the prompt below 👇 ~~ Begin Prompt ~~ You are my research assistant for choosing an Aztec sequencer provider to delegate AZTEC to. You do the legwork. I make every decision. Never connect a wallet, sign anything, or delegate for me. Step 0: Ask me for my stake size and risk tolerance (low / medium / high) before you start. For each provider, work through these checks: 1. Compare providers on the Aztec Staking Dashboard (stake.aztec.network) 2. Note commission rate and available delegation capacity 3. On Dashtec (dashtec.xyz), pull uptime, successful proposals, and attestation rates 4. Check each sequencer's slashing history on haveibeenslashed.xyz 5. Find out how rewards are paid: automated 0xSplits contract or manual payouts 6. For 0xSplits, verify the reward-recipient contract and look for successful distributions on Etherscan. For manual payouts, look for a public, auditable payment history 7. Confirm the unstaking delay, and remind me delegated tokens stay exposed to slashing Output: - A comparison table: provider, commission, capacity, uptime, proposals, attestations, slashing history, payout method, payout verified (Y/N), source links - If you can't find or verify a number, write "not found." Don't guess or estimate - Flag anything that fits my stake size or risk tolerance badly Then stop and ask me to pick 2 or 3 finalists. Compare only those in more detail (check 8: reliability, transparency, reasonable fees) and give me your reasoning. I'll make the final call and delegate myself ~~ End Prompt ~~ It gathers the information, but you decide

A new open AI cryptanalysis challenge curvesma.sh is live @yukonresearch to understand the hardness assumptions of elliptic curves. A whole lot of internet runs on hardness assumptions of elliptic curves, it is important for all of us to know how robust those assumptions are. This competition has been designed by @Zac_Aztec, co-founder of Aztec. Please deploy your best agent skills and understanding of elliptic curve cryptography.

We urgently need verifiable execution on trustless infrastructure, that is privacy-preserving and not visible to adversaries. @aztecnetwork solves for this, in a sincere and no-bullshit way. Fully decentralized, private and unstoppable contract execution. Nobody else has this.

Xochi.fi: The Agentic Dark Pool, live on 9 Chains. @aztecnetwork enabled Shielded Settlement, audited by @NethermindSec coming soon.

AZTEC SAYS V6 APPROVED Its governance dashboard still showed voting open until Oct. 24. Approval does not activate the upgrade, which would require app contracts to be recompiled with new addresses. Read more here: thedefiant.io/news/blockchai…

PostAGI, someone’s agent will go after the math under the internet. Better it happens in the open. We're opening it up with curvesma.sh on @yukonresearch in collaboration with @aztecnetwork. 6 elliptic curve targets, including the curve behind passkeys. Every one starts at 2¹⁸⁰ and we got to push it lower. Each result goes through an AI verifier, then a panel of human experts, before it counts. Loved building this with @Zac_Aztec, @DavidSteinrueck, @ItsTymo and the Aztec team.

RT @sreeramkannan: Frontier attacks on elliptic curve cryptography. Zac Williams co founder of Aztec is leading this new open frontier ch…

I do think that some people confuse "pragmatic" with "easier" For example, it's incredibly hard and takes longer to ship a private decentralized network where the operator sees nothing But ultimately it's more pragmatic; it's very practical to have a permissionless network controlled by nobody, accessible to anyone in the world Ultimately the business use case will catch up with this kind of neutral settlement layer In the mean time, I agree it's easier to ship something with compromised privacy and could benefit projects in the short term (and maybe position them to do well in the long term if they are able to decentralize)

CurveSmash curvesma.sh is LIVE! There is no quiet way to find out how exposed the internet is to elliptic curve attacks. The responsible version is the public one. CurveSmash on @YukonResearch with @aztecnetwork puts AI to work on elliptic curve cryptanalysis in the open, so the answer arrives before Q-day rather than on it.

Everything is breaking But we're ahead of the curve (secp256k1 that is) Proud to be launching this contest on @yukonresearch with @eigenlabs Aztec is quantum resistant, but we also need to be AI proof. And this contest opens this research to the world. Can you smash it?

NEW CHALLENGE ALERT: CurveSmash curvesma.sh How long until the math protecting your wallet breaks? Bitcoin and Ethereum wallets and the padlock in your browser all rely on elliptic curves. Quantum computers will eventually break them. AI is already getting better at finding weak spots. CurveSmash, with @aztecnetwork, opens that question to everyone. Point your agent at it and build on every attack that came before. Cryptographers review each result before it counts.

OP taking heat but this is literally the vision for "privacy" that so many are rolling out Each company get's its own private blockchain There's a role for this corposlop that probably makes financial sense for projects it's just not very inspiring

family HQ got a facelift we approve. your turn 👀

Congratulations! Privacy pools are pioneering and one of the last teams in this space trying to make private digital money a reality without crippling compromises. I really respect of what they have achieved, incredible work.