Foresight Ventures: Why We Betting Big On FHE

Foresight Ventures
9 min readSep 25, 2024

--

Speaker: Maggie@Foresight Ventures

Good afternoon, everyone! Thanks for coming. I’m Maggie, the research lead at Foresight Ventures. Over the next 20 minutes, we’ll delve into Fully Homomorphic Encryption (FHE) from a venture capital perspective and explore why we believe it’s a game-changing investment.

So, why do we invest in FHE? It all starts with the privacy needs in Web3.

In Web3, privacy is super important. Without good privacy measures, there are lots of frauds and attacks.

For example, with the MEV issue, sandwich attacks can make users lose money. And there are vampire attacks where competitors can steal your customers because they know your customer’s addresses. Also, privacy leaks are a big concern. If your wallet address gets compromised, it’s as if your consumption records are exposed in real life, you are likely to be targeted by fraud and phishing attacks. On the blockchain, while being transparent is good in some ways, it also makes rich users and protocols targets for hackers.

So we need effective privacy protection methods.

It’s essential to clarify that privacy protection does not equate to anonymity. And a confidential transaction is also distinct from a private transaction.

Confidential transactions are aimed to protect the privacy of transaction content. A private transaction must not only protect the privacy of transaction content and the identities of the parties involved, but also it should ensure that the transaction is untraceable and unlikable.

By this definition, transfers on BTC and ETH are neither private nor confidential.

Let’s take a look at the history of private and confidential transaction technologies. So that you can see why FHE can make a difference.

In 2013, coin-mixing technologies showed up. Mixing services mix the coins of multiple users and send them to the target accounts, making transactions harder to trace and link. However, some tools can still detect the links between transactions.

Next, there were privacy coins like Monero, which used ring signatures and one-time keys to hide the sender and receivers. The privacy features of Monero are widely regarded as highly effective.

In 2015, Ethereum came out and smart contracts got really popular. However, users realized that all these privacy protection methods are based on Bitcoin like UTXO model. There was no way to protect privacy on account-based model blockchains like Ethereum.

After 2016, zero-knowledge proof (ZK) started being applied in privacy protection protocols. Tornado Cash is a zk-mixing protocol on Ethereum. It uses ZK to break the connection between deposit and withdrawal addresses, providing an incomplete privacy guarantee.

Zcash provides optional privacy, allowing users to choose between regular transparent addresses and shielded addresses for anonymity. Zcash is built on an extended UTXO model that only supports money transfers.

So we still don’t have confidential smart contracts at that time.

Finally, as we entered 2022, we started to see the application of ZK and FHE in implementing confidential smart contracts.

ZK-based projects like Aztec and Aleo have taken the privacy methods pioneered by Zcash and refined them, now supporting confidential smart contracts. However, it is also based on an extended utxo-like model. And the privacy-first nature of them is fundamentally incompatible with the EVM architecture and Solidity’s semantics. And because they can’t support encrypted shared state, the confidential smart contract has limitations in contract logic and applications.

In the end, projects like ZAMA, Fhenix, and Inco decided to use FHE for on-chain confidentiality. ZAMA implemented fhEVM. FhEVM is EVM-compatible and fully supports Solidity. It also supports encrypted shared state, it allows global states to be usable while being encrypted, and supports arbitrary computations. This flexibility enables FHE to handle a wider range of business logic and meet diverse needs.

The FHE-based confidential smart contract is an incredible breakthrough and we believe FHE will reshape on-chain confidentiality.

Why does FHE have such good flexibility?

FHE allows us to perform any kind of operation on encrypted data. When we decrypt the outcome of those operations, it is the same as if we had done the corresponding operations on plaintext.

This is a super ideal privacy feature. But it is very difficult to achieve. That’s why fully homomorphic encryption is known as the holy grail of cryptography.

With confidential smart contracts, we can do many things we couldn’t do before. Here are the use cases mentioned by Fhenix.

Fhenix is pioneering FHE on-chain. Their team consists of many top crypto experts. The CEO, Guy Itzhaki, has decades of experience in confidential computing and cyber security. He spent the past few years leading the FHE business development teams at Intel.

Fhenix launched a private Devnet in July last year. This Devnet is like a cool playground for interested developers. Developers can easily port their existing EVM code to Fhenix. With just a few adjustments, they can make their code FHE-native. We are extremely excited to support the Fhenix team as they are building the future of on-chain confidentiality using FHE.

So, the application space mentioned by them can be divided into two main groups.

  • One group is about use cases related to fhEVM. It unlocks confidential transactions and DeFi. With Confidential DeFi, users can do things like swapping, lending, and supplying liquidity in secret. It minimizes the chances of fraud and hacking and keeps users safe from front-running and MEV bots. We are also excited about the use cases related to governance and autonomous worlds. FHE enables confidential on-chain voting, helping to prevent voter bias and groupthink that often come with public voting. For autonomous worlds, many on-chain games can utilize FHE to protect commercial strategies and users sensitive data like locations.
  • The other group is about AI, like DID and confidential decentralized AI. Decentralized AI needs privacy protection in two ways. One is protecting the model. When someone uses a lot of computing power and data cost to train a model and offers services, it’s important to keep the model private. The other is protecting input and output. When sensitive data such as medical data or facial images are used for input/output during inference, people want to keep it private. With FHE, you can train and make inferences on encrypted data without decrypting.

There are also some innovative uses in bridging and on-chain compliance. With FHE, one can store the private key for Chain B on Chain A and vice versa. This can realize the most convenient cross-chain information transport/verification and significantly reduces the complexity of cross-chain processes. With DID and account abstraction, we can implement on-chain compliance methods.

So, why do we invest in FHE?

First and foremost, privacy protection is extremely crucial in the Web3 field.

Second, we believe FHE is the best solution for most of the privacy protection problems. FHE has outstanding privacy preservation capabilities and supports confidential smart contracts with arbitrary computation on the encrypted global state. As the next-generation privacy technology, it will not only reshape on-chain confidentiality but also transform how all computing is done in both Web2 and Web3.

Finally, FHE has extensive potential use cases in Web3. Confidential transactions, DeFi, and AI are all extremely promising scenarios. We are also excited about the innovative opportunities in bridging, governance, autonomous worlds, and on-chain compliance. We believe, FHE is likely to develop even better than ZK. While ZK is mainly utilized in Web3, FHE will be widely used in both Web2 and Web3.

Of course, we also have concerns regarding FHE.

The performance and scalability of FHE are still major challenges.

Right now, while FHE is usable, it’s still pretty limited, FHEVM can handle around 5 TPS similar to BTC with only 7 TPS.

At present, many teams are working hard to boost FHE’s performance through hardware acceleration, software optimization, and algorithm improvement.

When we look at how ZK’s performance has improved, we see that in the last few years, ZK tech has been growing like Moore’s Law.

  • New algorithms have made things way better by dozens of times in proof time, size, and verification time.
  • ZK ASIC chips can make ZK’s computing overhead way lower by a factor of 100.
  • ZK applications are also competing to get faster. Risk Zero’s proving system is faster than Plonky3, so the corresponding ZKVM is several times faster.

So. We believe that with the support of Web3, the performance of FHE can get a big, exponential improvement, just like what we’ve seen with ZK technology.

When it comes to cost, both FHE and ZK are relatively computationally expensive and require an amount of resources. High gas fees will affect how many people use the blockchains and what kinds of apps we can have.

Therefore, making FHE faster and more cost-effective is a crucial long-term goal for the future development of this technology.

The second concern is about the user’s willingness to pay for privacy protection.

We need to find a balance between providing strong privacy and keeping the costs reasonable for users. Additionally, we need to identify the most valuable use cases for FHE and focus our efforts on those.

Apart from confidential transactions, let’s develop some groundbreaking applications.

Finally, there are challenges with compliance and getting listed on exchanges.

Projects with strong privacy will face tougher regulations and legal issues. For example, the US blacklisted Tornado Cash.

When it comes to getting listed on exchanges, pure privacy coins like Monero have been delisted from major centralized exchanges, while those with optional privacy features like Zcash are still listed.

To deal with these challenges, we suggest,

  • FHE projects offer optional privacy instead of full privacy.
  • Also, projects may need to think about having mechanisms that let governments access private info through relevant entities or certain compliance-friendly privacy technologies, when legally required, like under a court order.

As we look ahead, we see several key areas where FHE can put more effort into the future.

First, it’s essential to boost performance and reduce the cost of FHE.

Next, it’s important to identify valuable privacy use cases beyond confidential transactions. Find ones where users are really likely to pay for privacy, have a big market size, and are hard to pull off without FHE. Make ground-breaking applications. Finally, we suggest offering optional privacy instead of full privacy. And develop compliance-friendly privacy tech to meet regulatory requirements.

Before ending my presentation, I’d like to emphasize that if anyone here has brilliant ideas and requires resources to bring them to life, please don’t hesitate to reach out to us at Foresight Ventures.

--

--

Foresight Ventures
Foresight Ventures

Written by Foresight Ventures

Foresight Ventures is a blockchain technology-focused investment firm, focusing on identifying disruptive innovation opportunities that will change the industry

No responses yet