上轮牛市发动机是DeFi,下轮有可能和它相关?

白话区块链Pubblicato 2023-01-16Pubblicato ultima volta 2023-01-18

Introduzione

我们来说说ZK目前在各个赛道的发展状况和潜力。

这篇不是要讲Zk-Sync,Starknet,Scroll这些Layer2的,也不是讲Aleo,Aztec这些隐私公链的,这些全网已经说烂了,我们来说说ZK目前在各个赛道的发展状况和潜力。

ZK主要能拿来干什么

ZK是个今年被翻来覆去说的词,甚至有人把他提到了区块链第三重要的位置,第一个是比特币,发明了区块链这个概念,第二个是以太坊,引入了智能合约这个东西,第三个就是ZK。

ZK,零知识证明的定义也不累赘,但我想你从第一性原理出发,想象ZK可以在哪些场景下发挥作用

ZK最基本的两大特性:

证明了自己想证明的事情,同时透露给验证者的信息为"零"

生成证明很费劲,验证证明很容易

基于这两大特性,我们很自然的发展出了两大类应用:

1.隐私

2.扩容

基本上所有基于ZK的应用,都可以归到这两个大类里。

像Zcash,Aleo这种项目,毫无疑问是隐私类应用方向的C位,而Starkware,Zk-sync这种扩容类Layer2,毫无疑问是扩容类应用方向的C位。

那么C位之外,还有哪些场景呢?

隐私类拓展场景

Zcash或是Aleo主要是“我不想让人家知道我做了某笔转账或是执行了某笔交易”。

这是隐私最直接的场景。还有一些相关的场景,分别为身份、安全、游戏和机器学习,接下来分别说说:

1.身份

比如当你参加一个在线投票,你想要证明你投了但是保持匿名,又或是你想要申请一个贷款,你想要证明你的收入超过他们的要求,但又不想暴露你具体挣多少钱,又或是你要申请一个医疗补助,你想要证明你患有补助清单里的某一项疾病,但又不想透露具体是哪一项疾病……等等诸如此类,ZK都可以帮你隐去“部分你不想透露的信息”。

2.安全

比如有项目在做“链上保险箱”,你在存资产进去的时候,需要输入一个密码,而密码是基于ZK的,这样你即便私钥被盗了,资产也不会受影响。

3.游戏

不用多说,大名鼎鼎的黑暗森林就是把ZK融入了游戏,毕竟链上一切都是公开的,你要想实现星际争霸里面那种地图迷雾该咋办呢?上ZK啊!

4.机器学习

许多APP会收集你的各类信息,从睡眠质量到购买记录到每天走的步数……有些明着收有些暗着收,甚至还有给你奖励让你贡献信息的,他们收集信息的目的可能不是为了信息本身,而是通过信息的大数据去建模做机器学习,如果你想要奖励但是又不想暴露个人隐私怎么办?现在已经有基于ZK的机器学习项目了,数据我拿了,模型我也建了,但是收集到的跟你相关的个人数据我其实也不清楚,被ZK隐藏了……

扩容类拓展场景

扩容类C位毫无疑问是Starkware,Zk-sync,Scroll这几个,但其实扩容类周边的生态项目是很多的,其中有直接相关的,又间接相关的。

各举两个例子:

1.直接相关——性能加速

众所周知,ZK是个前景无比光明,但又处在非常早期的赛道。

早期的最大问题除了EVM兼容性,语言的适配性等各个问题之外,还有一个很大的问题就是性能。

想象ETH那让人诟病的TPS撑了这么多年,你肯定深有感触。

所以Zprice这种多个组织联合举办,奖金高达数百万美元的零知识证明加速大赛,从算法到硬件,各个团队在ZK加速上无所不用其极。

比如拿CKB当年的一张图来看,你就知道优化的空间有多大了,其中各个签名算法,证明方式等都在不断的演进,也都有优化的空间。

比如笔者知道的最常见的证明系统里面,Plonk就有优化的Hyperplonk版本,Virgo也有优化的Dvirgo版本。

还有目前有不少团队都在搞硬件加速,从GPU到FPGA到ASIC,都有团队在搞,三大解决方案当前也各有优劣,但按照以往比特币和以太坊的历史来看,通常是短期GPU+FGPA,最终还是会走到一个ASIC为王的终局。

2.间接相关

互操作性

目前业内有两个相对头疼的问题需要解决:

一个是ETH上面的L2越来越多,将来还会有L3,这么多L2,L3的流动性割裂和互操作性是个很大的问题。

还有就是ETH即便是老大,也不太可能一统江湖,总还是有其他Alt Layer1需要互通有无,尤其是Cosmos生态, IBC这么好的东西,只能内部生态使用,跟ETH那边无法对接,是让人很着急的事儿。

你可能说我们不是有桥么,那么多桥还不够你用的么?

可是桥去年损失的资金数量不要太多!而且目前的桥,最多也只是把IBC与ETH在资产上互相桥接下,不可能真的让ETH去享受类似原生IBC的“快乐”。

好在这两个问题都有团队在解决了!思路也都大同小异,就是用ZK来搞。

2023年,你肯定会看到ZKBridge,和ZK-IBC这类项目陆陆续续上线。

ZKBridge是相对通用的跨链桥,不再像之前那样需要相信一个第三方验证者的集合,只需要验证不同链上最后生成的ZK Proof就好,只是这个Proof生成的速度和体验,相信跟Zk-Rollup一样,都还有不小的优化空间,今年大概率体验不会“那么的好”。

而打通IBC和ETH则难度更大一些,之前IBC和ETH完全没法互通是因为IBC的轻客户端不能在ETH上跑,因为验证Cosmos那边签名消耗的Gas会以天量计。现在有了ZK,特别是递归证明的ZK,那么理论上就可以避开直接在以太坊上验证Cosmos那边的ed25519签名,而是构建一个签名有效性的zk证明,直接验证证明即可。

当然,这个ZK-IBC的难度比ZKBridge还要更大一些,所以可能会是2024年的事情了。

去中心化

还记得V神在Endgame里面说的那个关于去中心化的推断么?

区块生产难免不断的趋向中心化,而我们只要保证验证的去中心化即可。

ETH跑一个验证节点,需要32个ETH,和相对不低的硬件水平,虽说相对Solana等高速链来说门槛低了很多,但仍然不是随随便便一个电脑就可以完全胜任的。

假设有一天,ETH上面Fully Snarked ETH那条路线完全实现了,ZK Proof的生成速度提升到了10-15秒这个级别,也就是说ETH出一个块就能附带一个ZK证明。

那到时候,一切就都不一样了。

简单来说,ETH变成了一个类似Mina的存在。这时候,互操作性上了一个新的台阶,因为验证以太坊的区块变得无比容易。

更重要的是,去中心化也会上一个大的台阶,因为现在验证一个以太坊的区块,就不再需要什么“好一点的电脑”了,甚至理论上来说一部手机都可以轻松的验证以太坊,这对于“去中心化”这四个字来说,意义重大。

Letture associate

Earning 1000 Trillion in Half a Year, 'Pocketing' 20 Million per Capita: This Round of Wealth Creation in the Korean Stock Market is Unprecedented in Scale

The South Korean stock market is experiencing an unprecedented wealth surge in 2026, with household equity and fund asset values soaring by over 1,000 trillion KRW (~$730bn) year-to-date. This translates to an average per capita wealth increase of roughly 20 million KRW, fueled by a historic 109% rally in the KOSPI index. The boom is driven by three converging forces: an AI-driven semiconductor supercycle boosting giants like Samsung and SK Hynix; the government's "Value-Up" market reforms addressing long-standing corporate governance issues; and aggressive real estate regulations that have locked capital within financial markets, preventing profits from flowing back into property. This has triggered a wealth effect, boosting high-end consumption significantly. However, the gains are highly concentrated. The two semiconductor behemoths account for over half the index's value, but retail investors own relatively low stakes in them, systematically missing the biggest rallies. Wealth and consumption benefits are skewed towards luxury goods and imported cars, bypassing mainstream retail. Further risks stem from excessive leverage, with high trading volume in leveraged ETFs, and a market sentiment heavily reliant on the AI sector's fortunes and speculative rumors. While this cycle marks a potential shift from real estate to equities as a primary wealth generator for Koreans, its sustainability, amid structural imbalances and leverage, remains a critical test.

marsbit9 min fa

Earning 1000 Trillion in Half a Year, 'Pocketing' 20 Million per Capita: This Round of Wealth Creation in the Korean Stock Market is Unprecedented in Scale

marsbit9 min fa

imToken's 10th Anniversary Unveils Strategic Direction for the Next Decade: Evolving from a Trusted Main Wallet to a Personal Digital Hub

On its tenth anniversary, decentralized wallet imToken announced its strategic vision for the next decade: evolving from a "trusted main wallet" into a "personal control interface." This new direction aims to help users manage not only digital assets but also identity, permissions, and AI agent actions in an increasingly open and intelligent internet. imToken outlined that while the past decade focused on Store, Send, and Stake—securing assets, enabling transfers, and facilitating network participation—the future introduces a fourth core proposition: Sign. This expanded concept goes beyond transaction signing to encompass expressing intent, granting permissions, setting rules, delegating actions, and revoking authorizations. As AI agents gain autonomy, imToken emphasizes the need for clear, verifiable, and revocable user control over their actions. CEO Ben He stated that imToken's mission is shifting from enabling ownership of digital assets to ensuring user sovereignty over their entire digital world in the AI era. The company's core principle has been upgraded from "Digital Assets, Under Your Control" to "Your Digital World, Under Your Control." Future development will focus on three areas: upholding self-custody principles, extending security from transactions to authorizations and automated actions, and building product capabilities for managing permissions, delegations, policies, and revocations. imToken views the wallet's role as expanding into a trusted control interface for human-AI collaboration, where managing keys, signatures, and permissions forms the infrastructure for personal digital sovereignty. Founded in 2016, imToken serves millions of users across 150+ countries, providing non-custodial wallet services supporting over 50 blockchain networks.

marsbit12 min fa

imToken's 10th Anniversary Unveils Strategic Direction for the Next Decade: Evolving from a Trusted Main Wallet to a Personal Digital Hub

marsbit12 min fa

With 300 Million Financing to Accumulate ETH, the Hidden Concerns Behind BitMine's High-Yield Preferred Shares

BitMine, led by Thomas Lee, plans to raise up to $300 million through an initial public offering of 3 million shares of perpetual Series A preferred stock on the NYSE (ticker: BMNP). The stock offers a fixed 9.5% annual dividend. The funds are intended to further the company's accumulation of Ethereum, expand its staking node operations, and for general corporate purposes. This move comes as BitMine faces significant challenges. Its massive Ethereum holdings, over 5.3 million ETH (roughly 4.5% of circulating supply), are currently at an unrealized loss exceeding $8.5 billion due to the crypto market downturn. The company's core business model relies on staking these ETH holdings to generate yield, which it presents as the primary means to cover the new, substantial annual dividend obligation of approximately $28.5 million if the offering is fully subscribed. While the model is similar to MicroStrategy's bitcoin-focused strategy of using capital markets to fund crypto acquisitions, BitMine's product differs with its fixed, non-adjustable dividend rate. The company acknowledges risks, stating dividend payments could also come from cash reserves, asset sales, or future financing, and warns that staking yields may underperform or be illiquid during market stress. The 9.5% fixed rate reflects the higher risk premium demanded from investors for a company heavily exposed to Ethereum's volatility.

Foresight News15 min fa

With 300 Million Financing to Accumulate ETH, the Hidden Concerns Behind BitMine's High-Yield Preferred Shares

Foresight News15 min fa

A Year of Observing Agent Payments: The Cold Reality Behind the Hot Narrative

A Year in Agent Payments: The Cold Reality Behind a Hot Narrative This article examines the current state of "Agent payments," a year after it became a major trend at the intersection of AI, payments, and crypto. Despite significant investments from major players like Stripe, Visa, and Google, the author—having built products and spoken with merchants and developers—finds genuine, large-scale demand still lacking. Key findings across several hyped scenarios reveal structural challenges: * **Agent-to-Merchant Commerce:** For most product categories (e.g., clothing, electronics), AI shopping via chat is inferior to traditional visual e-commerce. Merchant interest is largely defensive, focused on future-proofing rather than current consumer demand. True potential exists only in specific, high-frequency/low-decision scenarios (like food orders) or for simplifying broken checkout experiences, but these require massive consumer distribution, favoring incumbents. * **Agent-to-API/Machine Commerce:** While stablecoin micropayments are touted for API calls, developers already solve small-value payments via prepaid credits and subscriptions. Large SaaS providers prefer enterprise contracts over fragmented micro-pricing. The market exists for long-tail services outside the top providers but is inherently smaller than the hype suggests. * **Agent-to-Agent Payments:** This remains a theoretical long-term vision with negligible real transaction volume. The core challenges—discovery, trust, negotiation, dispute resolution—are unsolved. While the potential for a new, high-speed settlement layer is real, it is not the current market. * **Agent Finance:** This is the sole area with existing, paying customers (fund managers, DeFi users). AI enhances real-time monitoring and autonomous rebalancing, offering real capability gains. However, competition favors established, regulated institutions with existing licenses and client relationships. The author concludes that the core deficiency in the Agent economy is not merely a payment layer, but a more complex **coordination** capability—figuring out how Agents and humans work together, verify task completion, and settle outcomes. Payment is just one component of settlement, which is itself part of coordination. For large companies, investing now is a defensive, long-term bet with minimal cost. For startups, however, the imperative is to find markets that exist today, not wait for a future wave that remains on the horizon.

marsbit17 min fa

A Year of Observing Agent Payments: The Cold Reality Behind the Hot Narrative

marsbit17 min fa

Trading

Spot
Futures
活动图片