ZK on the rise, an anticipated capital game

Written by: 0x5willows, Deep Tide TechFlow

In the movie "Let the Bullets Fly," there is a scene where in order to convince others that he did not eat an extra bowl of cold noodles, Liuzi cuts open his belly, proving his innocence at the cost of his life.

Here, Liuzi is the prover, the onlookers are the verifiers, and Liuzi's method of proof is to show whether there is indeed an extra bowl of cold noodles in his stomach. However, the cost of this method of proof is Liuzi's life.

The above example is a typical proof problem.

First, to ensure the validity of the verification, the prover must share knowledge (in the above example, the situation inside Liuzi's stomach corresponds to the knowledge). If the prover wants to pass the verification, they need to inform the verifier of the knowledge, and informing knowledge often comes at a cost.

Second, if the verification is invalid, the verifier will face the risk of fraud. For example, in a courtroom, if the defendant can obtain all the questions from the prosecution's lawyer before the trial, they are very likely to successfully fabricate a perfect story to deceive the other party.

The initial solution was to introduce a third party to make the verification process public, thus reconciling the conflict between knowledge ownership and usage rights. However, even so, for exclusive knowledge (such as passwords), once the usage rights are relinquished, the ownership is also lost. Is there a verification method that does not require the use of knowledge?

The Birth of Zero-Knowledge Proofs

In 1985, Dr. S. Goldwasser graduated and came to MIT, co-authoring a landmark paper titled "Knowledge Complexity in Interactive Proof Systems" with S. Micali and Rackoff, which introduced zero-knowledge proofs.

Since then, the proof process has theoretically supported the idea that it does not require "knowledge." In simple terms, through zero-knowledge proofs, one can keep secrets while still convincing others.

However, due to efficiency and applicability shortcomings, zero-knowledge proofs remained at the academic theoretical level for a long time or could only be used for specific projects until they encountered blockchain.

The two key concepts of zero-knowledge proofs are: keeping secrets, i.e., "not leaking information"; and convincing others, i.e., "proving the validity of assertions."

These two characteristics are precisely what blockchain needs:

Privacy: The consensus requirement of blockchain networks demands everything to be open and transparent, making privacy protection a problem.

In privacy scenarios, leveraging the "not leaking information" characteristic of zero-knowledge proofs allows for the proof of valid asset transfers on the blockchain without disclosing transaction details (recipient, sender, transaction balance).

Scalability: The decentralized nature of blockchain makes verification an unbearable burden, and lightweight proofs become a necessity.

In scalability scenarios, the focus is mainly on the "proving the validity of assertions" characteristic. On-chain resources are limited, so we need to migrate a large amount of computation off-chain, and zero-knowledge proofs can effectively prove that these off-chain actions are trustworthy.

As early as 2016, ZCash, which focuses on on-chain privacy protection, officially launched, completing transaction verification through zero-knowledge proofs without disclosing all transaction information (sender, recipient, transaction amount). However, after years of development, ZCash remains lukewarm.

The reason is simple: although privacy is important, current blockchain users have not yet established a widespread and strong awareness of privacy. Are they afraid to show off their wealth or their poverty?

Currently, privacy is not a pressing need; scalability is, especially for Ethereum.

The Combination of ZK and Rollup

From the perspective of underlying protocols, the cost of on-chain transactions is inevitably high because sufficient decentralization and security require a sufficient number of nodes for repeated verification.

As blockchain applications expand, the time to synchronize a node is getting longer, and the increasing on-chain data volume raises hardware requirements. Most home computers cannot even meet the basic requirements for nodes. Public chains like Ethereum, which have a wide range of uses, have only about ten thousand nodes, most of which are hosted on Amazon AWS, contradicting the original intention of decentralization.

Compared to privacy protection, the scalability of blockchain is clearly more meaningful.

As long as scalability is achieved, it can maintain the "politically correct" decentralization while reducing gas fees.

In August 2017, Ethereum co-founder Vitalik Buterin and Joseph Poon jointly proposed the first-generation scalability solution, Plasma.

However, Plasma could not provide the same data availability and security as the main chain and quickly faced a complete collapse, leading developers to turn their attention to Rollup technology.

The core idea of Rollup is quite simple: it is to package a large amount of transaction data originally scattered in blocks into "condensed" transactions and publish them on-chain. To ensure the validity of each transaction, various Rollup schemes have designed different mechanisms to ensure the entire process's security remains consistent with Layer 1.

In this direction, there are mainly two schemes: ZK Rollup and Optimistic Rollup. The former ensures security through cryptographic techniques of zero-knowledge proofs (ZK-SNARKs), while the latter inherits the penalty mechanism from Plasma, where nodes that act maliciously will pay a heavy price.

On September 1, 2021, the Ethereum scalability network Arbitrum, based on Optimistic Rollup, announced that the public test version of the mainnet was officially launched, marking that OPR has taken the lead over ZKR and officially stepped onto the stage.

Compared to the OPR scheme relying on penalty mechanisms, the ZKR, based on mathematics and cryptography, can clearly achieve a higher level of trustlessness. However, the first technical challenge that Rollup must overcome is the compatibility issue with EVM.

If we consider EVM as a computer, it calculates the output results of smart contract operations given specific inputs. Optimistic Rollup solutions, including Arbitrum and Optimism, have EVM-compatible virtual machines that allow them to handle all operations occurring on the Ethereum main chain.

In contrast, ZK Rollup faces a challenge because, at the beginning of EVM's design, developers fully anticipated the potential use of ZK technology, making ZK-EVM the first hurdle that ZK Rollup needs to overcome.

Developers face two choices: design a ZK instruction set that supports the existing EVM or redesign a virtual machine that is friendly to zero-knowledge proofs. The former route is represented by Hermez and the Ethereum Foundation's Applied ZKP, while the latter is mainly represented by zkSync and Sin7Y.

Outside the battlefield of L2, ZK-EVM has become a small battlefield where major tech teams compete.

In August 2021, Polygon (Matic) acquired Hermez Network, which is dedicated to developing ZK-EVM, for $250 million.

In April of this year, zkRollup Scroll, focusing on zkEVM solutions, announced the completion of a $30 million Series A financing round. The Scroll team will collaborate with the Ethereum Foundation's Applied ZKP team to launch a zkEVM at the bytecode level and explore hardware to accelerate ZK proof generation, building a decentralized proof system.

Expectations Run High

There is a consensus among leading capital that ZK is a definitive technology that may become ubiquitous in the blockchain world in the future.

As major institutions like Paradigm, a16z, and Sequoia intensify their investments in ZK, the FOMO sentiment around ZK in the market has been ignited, and the valuations of ZK projects in the primary market are gradually moving towards dream rates.

In January 2018, the ZK-Rollup development team StarkWare raised $6 million in seed funding, with investors including Ethereum founder Vitalik Buterin, Paradigm, Pantera, PolyChain, and other star institutions.

In November 2021, StarkWare raised $50 million in Series C funding, reaching a valuation of $2 billion, led by Sequoia Capital, with participation from Paradigm, Three Arrows Capital, Alameda Research, and others.

In March 2022, when StarkWare raised a new round of financing, the valuation reported in the primary market had reached $6 billion.

Another ZK star, zkSync's parent company Matter Labs, completed a $50 million Series B financing round last November, led by a16z, with Placeholder, Dragonfly, 1kx, and others participating.

Due to the fact that there is still a lot of theory in the ZK field that has not been put into practice, and the lack of successful cases and quality codebases as references, the learning curve is very steep, and the development difficulty of projects is extremely high. The high threshold undoubtedly increases the value of ZK projects, but the current ZK track still carries the risk of overheating.

Investors who have been paying attention to the ZK market, like Evans, told Deep Tide TechFlow, this is a situation where expectations are running high. Currently, many public chains have yet to break through StarkWare's FDV (Fully Diluted Valuation). Achieving such a valuation without a concrete ZK ecosystem is essentially the result of short-term expectations being pushed in, and it may be challenging to maintain such a valuation going forward.

Another investor noted that a certain ZK project had increased from an initial valuation of $40 million to $400 million in less than a year, exclaiming, "Dream rate, can't afford to invest," even if it was just a DEX in the ZK ecosystem, some primary market valuations were also above $200 million.

However, the pursuit of ZK in the primary market is not necessarily a bad thing. With the backing of capital, a large number of excellent developers are investing in related research, accelerating the landing of EVM compatibility.

Taking ZKSync 1.0 as an example, most people, including Ethereum Foundation researcher Justine Drake, believe that achieving ZKSync 2.0 would at least require waiting until the end of 2022. However, in February of this year, the ZKSync 2.0 test was officially launched, becoming the first EVM-compatible ZK Rollup on the Ethereum testnet.

Nevertheless, above ZK Rollup, clouds of liquidity fragmentation, reduced composability, and centralization risks still loom. ZK has enormous potential, but it is far from being widely applied. Currently, what is referred to as ZK in the market more often points to ZK Rollup.

Essentially, the zero-knowledge proofs in ZK Rollup only outsource verifiable computation, which means providing a proof of computational integrity through a third party. While ZK Rollup alleviates the shortcomings of L1 scalability, it also faces pressure from computational costs.

Currently, there is no dedicated hardware for zero-knowledge proofs on the market. In Ethereum's roadmap, ZKEVM will be integrated in the future, and miners will need to generate proofs, which will require a chip capable of quickly generating zero-knowledge proofs.

According to Paradigm's predictions, the future market size of "ZK miners" is expected to rival that of the PoW mining market, with FPGA being the most important technology for accelerating ZK hardware. The cost of GPUs is too high, and their energy consumption is excessive. ASICs typically require 12 to 18 months or longer from design to manufacturing to deployment. In contrast, the FPGA supply chain is more lightweight and flexible.

Those who sense the "investment or speculative expectations" have begun to lay out their strategies. For example, a certain "crypto elder" who always stands at the forefront has started to engage in ZK hardware projects; some VCs are also trying to encourage individuals with hardware startup experience to undertake ZK hardware projects and incubate them personally…

From this perspective, even if ZK only exists as Rollup, its potential is far from being exhausted. Furthermore, for ZK to break through the narrative framework of Rollup and expand its use cases in privacy and trust, it may still need to wait for the rise of personal sovereignty awareness in the Web3 world.

Overall, ZK is a definitive technology whose significance even transcends blockchain. The crypto industry, which has high hopes for ZK, has already invested billions of dollars, looking forward to ZK helping blockchain undergo transformation.

If ZK development encounters obstacles or is disproven, the development of blockchain will once again experience a "narrative collapse" winter. This is a capital gamble where "success is the only option, failure is not allowed."

ZK, rise up!