CEO @ StarkWare | 2xCo-founder: Zcash, StarkWare ZK, Blockchain: ZK-STARK, Zcash co-inventor Opinions mine, no investment advice blockchainthebook.com

Israel
Yahoo! This is an exciting day for me. After a long journey through blockchains, cryptography, and ZK rabbit holes, my new book, co-authored with @NathanOnCrypto, is finally out. It's part history, part deep dive, part red pill, and it's about two of my favorite topics: blockchains and ZK. This is a book for non-experts, written in plain language, zero complicated jargon. And now it's finally published! Grab a copy for yourself or send one to a friend, and start the journey 👇
59
45
389
64,108
Here's the Big Question for Mathematicians with respect to AI: If you give AI a clear problem to solve, it's probably better than all human mathematicians (including Terry Tao) at solving it. But what about asking the questions? An example: In the episode below, Shafi Goldwasser, co-inventor of ZK, told me how ZK came about: they tried to solve the problem of playing poker over the phone or tossing a coin and agreeing on the outcome even if both counter-parties don't trust each other. Or take my own discovery of the FRI and ZK-STARK protocols -- by trying to make PCPs into succinct proofs. So the question is: Will AI be better than humans at picking the right questions to ask and prioritizing them well, and why? To be honest, I don't have an answer yet nitter.net/StarkWareLtd/status/21…
How do you prove something without revealing it? Eli Ben-Sasson meets ZK co-inventor Shafi Goldwasser on Block Smart to explore her Eureka moment, ZK beyond blockchains, and what it could unlock for AI. Chapters: 0:00 Greetings and intro: Shafi Goldwasser 1:22 ZK for non-mathematicians 5:01 How Zero-Knowledge came about 8:53 What did you imagine ZK would be used for? 13:03 Beyond blockchain: where else can ZK go? 14:58 Why ZK fits blockchain? 19:11 The case for trustlessness 20:29 AI meets ZK 24:02 The Eureka moment 27:35 Quick-fire questions 29:40 Parting thoughts
10
8
46
3,820
Spectacular result! combination of AI and HI (H for human). Congrats! Question to @Pooyahat and everyone: 5 years out, do math breakthroughs require HI?
New paper with Eshan Chattopadhyay, Chin Ho Lee, Shachar Lovett, Avishay Tal and Emanuele Viola: Exponential correlation bounds for polynomials and polylog-seed PRGs for AC0[2] eccc.weizmann.ac.il/report/2…
5
2
23
3,978
I love this direction, our original intent with Zerocash paper (the whitepaper preceding Zcash) was to bring privacy to Bitcoin. Haven't read this one, but would love to see our vision of privacy and scalability via ZK materialize on Bitcoin L1.
Shielded Bitcoin: Private Transfers on Bitcoin L1
28
26
200
14,486
ZK, ZK, ZK, and more ZK. Had the most interesting conversation with the legendary Shafi Goldwasser. Watch, comment, share your thoughts. (links to Spotify and Youtube in the next tweet)
How do you prove something without revealing it? Eli Ben-Sasson meets ZK co-inventor Shafi Goldwasser on Block Smart to explore her Eureka moment, ZK beyond blockchains, and what it could unlock for AI. Chapters: 0:00 Greetings and intro: Shafi Goldwasser 1:22 ZK for non-mathematicians 5:01 How Zero-Knowledge came about 8:53 What did you imagine ZK would be used for? 13:03 Beyond blockchain: where else can ZK go? 14:58 Why ZK fits blockchain? 19:11 The case for trustlessness 20:29 AI meets ZK 24:02 The Eureka moment 27:35 Quick-fire questions 29:40 Parting thoughts
10
18
93
6,410
Why do I feel on track to my $5K EoY zcash:native prediction? More and more whales asking me: Why price surge? Me: No fucking clue. What's your explanation? Not investment advice, DYOR.
Putting money where Mouth: I predicted that ZEC > $1200 by Sept. 25. Result: I was right 😎 Now, time for EoY prediction: ZEC > $5000. Let's see if I'm right again. (I'm not in it for the 27X reward, I just like being right) Not investment advice, DYOR.
46
31
279
29,949
Get ready for the Vulcan Economy: Remember Mr. Spock from StarTrek? The rational smart right hand of Captain Kirk? With AI, each of us will use a Mr. Spock for our commerce. Mr. Spock isn't moved by rage, emotions, shouty ads. He's super-rational. Build your biz for Mr. Spock, or go belly up.

ALT Star Trek Vulcan GIF

15
3
39
3,130
2 features that will become a baseline requirement for crypto: - Privacy - Post-quantum security Both will have to be easy to use and at a low cost.
28
21
185
20,272
Funny: I'm Kindle-reading Blood Meridian by Cormac McCarthy. Half of the words I look up are unknown to dictionary. Side notes: (a) 3rd time I'm reading of it (b) Its a crime he didn't get the Nobel for Literature!
11
1
41
3,063
Eli Ben-Sasson | Starknet.io retweeted
Fun fact about the Zerocash paper: It won the 10-year mark test of time award in 2024!
12
17
109
5,110
This is what I've been dreaming about for more than a decade: As Zcash co-founder I've been saying for more than a decade that the real problem of privacy is making it usable with good UX for end-users, for doing more than just transferring a single asset (for that case, ZEC rocks!). Great job @sendmoodz @EkuboProtocol !
You can now swap privately via Ekubo Protocol on Starknet! Try it out here: ekubo.org/swap?inputCurrency…
35
37
249
26,020
People don't realize the importance of a **seamless** migration to quantum-safety. At least, not yet. Without a seamless migration scheme, people will not upgrade their quantum-vulnerable accounts to quantum-resistant ones. This will be one of the most critical factors for quantum-readiness.
17
15
103
5,344
My prediction: ZEC flips SOL before SOL flips ETH
🚨 JUST IN: Multicoin Capital co-founder Kyle Samani predicts Solana will flip Ethereum this market cycle, claiming “no one really uses Ethereum” today.
64
36
472
41,670
Will Artificial Super Intelligence exterminate humanity? Those worried about ASI say that when it arrives, its advantage over us humans will be like our advantage over, say, mice. And then, it'll exterminate us just like we exterminated so many dumber species. The Most Important Question: If we train AI on human intelligence outputs, can it outsmart and transcend the intelligence of humans? If the answer is no -- it means we won't get ASI. If the answer is yes -- higher chance we're doomed. What's the evidence for transcendence of AI beyond what its trained on?
23
3
44
5,284
DYOR (ask your favorite AI), not investment advice.
whats the next Zcash zcash:native and why? i asked my GPT-6 Astra xhigh. the clanker says ethereum:0xca14007eff0db1f8135f4c25b34de49ab0d42766 chat, is this agi?
15
15
134
15,226
As Zcash co-founder I’ll say: Privacy is important for DeFi:
Shielding what people are already doing is the most important battle for onchain consumer apps. There is nothing more important than that. The last PvE.
33
25
226
13,343
Some folks (Bitcoiners) won't like this, but I have to say it. Part of the rise in ZEC is due, I think, to the fear that Bitcoin has become too ossified. I say this as someone who loves and cares deeply about Bitcoin. There, I said it.
81
49
602
41,433
Wow, very interesting assessment, thx!
很久不见,在众多曾经风光无限的 L2 公链中,@Starknet 算是走出了一条属于自己的路 一句话概括方向 Starknet 已经从单一的“以太坊 ZK 扩容层” 重新定位成:扩容以太坊、保护隐私、把量子安全的比特币带上链。 7 月 21 日的推文,三个关键词,恰好对应了它这一年多的三步走: 2025 年让比特币变得有生产力 2026 年让比特币变得私密 2027 年让比特币在量子时代变得持久 而且每一步都能在官方推文里找到对应的产品落地。 🔸起点(5 月),先解决隐私。 5 月 25 日那条:你在链上做的一切都是永久公开的,直到现在,是这一整年叙事的开端。同月引入 strkBTC,紧接着 STRK20 隐私标准立项,把隐私模型从 BTC 扩展到 ETH、STRK、USDC、RWA 甚至 meme 币,任何资产,一键隐私。 6 月还上线了一个挺有意思的小工具:钱包隐私分数"(score.starknet.io),让你贴地址就能看自己暴露程度,属于用产品化方式做用户教育。 🔸中段(6–8 月),把比特币变成能干活的资产。 6 月 23 日 strkBTC 正式成型:用自己的钱包把 BTC 跨到 Starknet,在 BTCFi 里用,再屏蔽进私有 DeFi。 8 月 6 日的一条长推文把机制讲透了:桥进来的 BTC 锁在一个由五家独立机构组成的联邦多签里,1:1 铸出 strkBTC,可随时赎回原生 BTC,没有任何单一方能冻结你的仓位。 🔸近段(8–9 月),量子安全落地。 8 月 27 日是个标志性节点:全球第一笔量子安全比特币交易在主网被挖出。 到 9 月,他们把这件事变成了开放竞赛(联合 StarkWare、Yukon Research、Eigen Labs),因为生成这笔交易花了 3100 GPU 小时、约 320 美元算力,他们公开悬赏两万多美元让社区把成本打下来。 费这么大周折,很多铁子好奇,strkBTC 和比特币在 Starknet DeFi 里的用途是什么 官方那条 “What can you do with strkBTC” 的推文基本给了标准答案:桥进来、然后赚、交、借贷、做市。 拆开来看: 🔸借贷。 最亮眼的例子是 7 月 22 日的 Sats Terminal 的说明:可以用 BTC 抵押借出 USDC。当时甚至出现过净年化 -2.04% 的 BTC 贷款,也就是说你抵押仓位赚的比借款成本还高,出现了负利率。 这种极端行情能出现,本身说明 Starknet 上的 BTCFi 流动性和收益结构已经跑起来了。 🔸质押。 这是我认为最有分量的一块。7 月 29 日他们上线了 DeFi 里“第一个比特币的私有流动性质押”。 机制很巧妙:收益来自 Starknet 把一部分 STRK 增发分给 BTC 质押者,Endur 自动把 STRK 换回 strkBTC 累积进你的仓位,你最终拿到的是比特币计价的收益。而且可以直接从隐私池里质押 “屏蔽状态” 的 strkBTC,余额数字不变、可赎回价值上涨。 🔸交易 / 聚合策略。 通过 @avnu_fi 可以直接 swap 进 strkBTC,Midas 则把散落在多个 app 的 BTC 收益策略聚合到一个入口,主打“比特币不该为了赚收益去开五个 app”。 跟其他 L2产品不一样的地方还有隐私和屏蔽能力 核心逻辑他们讲得很清楚,隐私是比特币进入 DeFi 缺失的那块拼图。技术上依托 STARK 证明本身 “后量子安全” 这一底层特性,产品上做成了 STRK20 机密标准 + 隐私池(privacy pool)。 Starknet 的做法是 shield/unshield:你的资产以加密票据形式待在隐私池里,质押时由池子直接花费这张票据,不用暴露、不用离开就能继续在 DeFi 里干活。9 月那篇《Crypto cannot run an economy in public》文章把这点提到了“公开的经济体系跑不起来”的高度。 坦白说,这个方向我是看好的 也挺佩服 @Starknet 不断折腾的韧劲,比特币 L2 一大堆,隐私协议也一大堆,但“把比特币变成私密且有收益的 DeFi 资产、同时铺量子安全的地基”这个交叉点,几乎是 Starknet 独占了。 希望明年回看这条时间线时,strkBTC 能被更多人看到
6
5
48
8,742
Fun* fact: The STARKs paper got rejected 4 times until it found its way to acceptance. * fun might not be the right word here ;-)
12
4
103
5,964