Multicoin founder talks about leading the investment in Zama: ZKP is the past, the future belongs to FHE

Original Author: Kyle Samani, Founder of Multicoin Capital

Translator: Azuma, Odaily Planet Daily

Editor’s Note:

On March 7, the open-source cryptography company Zama, aimed at helping developers build privacy-preserving applications, announced the completion of a $73 million Series A funding round, co-led by Multicoin Capital and Protocol Labs, with participation from Metaplanet, Blockchange Ventures, Vsquared Ventures, Stake Capital, and several founders of other projects including Juan Benet (Filecoin), Anatoly Yakovenko (Solana), and Gavin Wood (co-founder of Ethereum and Polkadot).

In the early hours of March 8, Kyle Samani, founder of Multicoin Capital, wrote an article titled "The Holy Grail of Cryptography," in which he elaborated on the rationale behind Multicoin Capital's investment in Zama and predicted the potential application value of Zama and its promoted "Fully Homomorphic Encryption" (FHE) in the cryptocurrency industry.

Below is the full text by Kyle, translated by Odaily Planet Daily.

In a sense, every server-side security breach (or attack) that has occurred over the past 30 years shares the same fundamental cause—certain sensitive data was not encrypted, or the attacker obtained the keys used for decryption.

When we describe the problem this way, the solution becomes clear—always keep data encrypted and ensure that decryption keys are never stored on any server.

The issue with this solution is that it was once impossible to achieve… but that is no longer the case.

Today, I am excited to announce that Multicoin Capital, together with Protocol Labs, has co-led a $73 million investment in Zama, with other participating institutions including Metaplanet, Blockchange, VSquared, Stake Capital, and Portal Ventures, as well as several other project founders including Juan Benet (Filecoin), Gavin Wood (Polkadot), Anatoly Yakovenko (Solana), Julien Bouteloup (StakeDAO), and Tarun Chitra (Gauntlet).

Zama is a leader in the "Fully Homomorphic Encryption" (FHE) space. FHE is characterized by its ability to perform arbitrary computations on encrypted data. The concept has existed in academia for decades but has not been practically valuable due to excessive computational intensity.

The Zama team, consisting of over 30 PhDs and cryptography experts, has achieved dozens of breakthroughs around FHE over the past four years. For certain types of computational work (primarily in the blockchain space), Zama's TFHE-rs and Concrete open-source libraries are ready for production. In the coming years, we expect to achieve a 1000-fold performance improvement in mathematics, software, and hardware (such as ASICs), enabling FHE-based encrypted computations to compete directly with plaintext computations.

Odaily Planet Daily Note: The following is an example of how FHE can be used in the field of privacy computing.

Assume we have an encryption function f that can transform plaintext A into ciphertext A' and plaintext B into ciphertext B', meaning f(A) = A' and f(B) = B'; additionally, we have a decryption function f' that can decrypt the ciphertext back into the plaintext, meaning it can convert ciphertext A' back into plaintext A and ciphertext B' back into plaintext B.

For general encryption functions, if we add A' and B' to get C', decrypting C' using the f' function will only yield a meaningless string of garbled text.

However, if f is a homomorphic encryption function, then decrypting C' using f' will yield the result C, which will also equal the sum of A and B.

In this way, the rights to process data and the ownership of data can be separated, thus preventing data leakage without affecting computation.

I believe that it won't be long before every Web2 application, from database applications to large language models, will incorporate FHE technology.

FHE will fundamentally change the underlying structure of computation.

One major characteristic of the blockchain space is that it can create immense financial value even with relatively limited transaction throughput. Therefore, given the current performance limitations of FHE, Zama will initially focus on the blockchain sector.

For Multicoin Capital and Protocol Labs, this is the primary reason we led the investment in Zama and aim to bring its technology to market. We have been closely collaborating with the Zama team to help them implement FHE on-chain.

Regarding on-chain privacy, there are currently many solutions available in the market. For simple payments, the effectiveness of zero-knowledge proofs (ZKP) has been well established; however, for transactions involving shared states—this nearly covers all DeFi transactions—ZKP often suffers from insufficient utility or extreme difficulty in implementation.

The beauty of FHE lies in its ability to allow developers to execute the same logic under an FHE-encrypted environment as they would in a transparent environment.

Currently, three teams have announced plans to build FHE-based blockchains using Zama's technology, including Fhenix (which we have also invested in), Inco, and Shiba, all of which will launch this year.

Zama can provide out-of-the-box fhEVM for developers looking to implement FHE on EVM, and I expect that Zama will soon support more blockchain virtual machines beyond EVM.

The introduction of FHE will enable new possibilities in several areas:

• Privacy-focused DeFi applications;

• Encrypted voting;

• Privacy-oriented DataDAOs;

• On-chain blind auctions;

• On-chain gaming;

• A new cross-chain thinking approach (storing FHE-encrypted private keys of chain B on chain A, and vice versa);

We are still in the early stages of exploring use cases, and I am very much looking forward to seeing developers leverage FHE to build innovative new applications.

In 5 to 10 years, we may look back at some of the current cloud data breaches in shock—people will be unable to understand why user data was not encrypted.

A few years ago, the Zama team proposed their vision for iterating HTTPS, calling it HTTPZ. HTTPZ is the natural evolution of internet encryption logic, where users encrypt data locally with their own keys and then send the encrypted data to the server, which can then compute on this encrypted data.

In the configuration of HTTPZ, it can be assumed that the server is always at risk of intrusion, but if all user data is encrypted and the keys are kept by the users on their devices, then even if an attacker accesses the server, it would be futile.

HTTPZ represents the grand vision of applying encryption technology to the internet.

Based on the trend of cryptographic technology gradually becoming mainstream over the past 30 years, we believe that this day will eventually come; it is just a matter of time.

We are incredibly excited to support Rand, Pascal (co-founder of Zama), and the entire Zama team as they lead the revolutionary wave of cryptographic technology.