FHE: Next Generation Privacy Protection Technology

FHE (Fully Homomorphic Encryption) is an advanced encryption technology that allows computations to be performed directly on encrypted data, thereby processing data while protecting privacy. This technology has a wide range of application prospects in fields such as finance, healthcare, cloud computing, and machine learning. However, due to its significant computational overhead, the commercialization of FHE still faces challenges.

Basic Principles of FHE

The core idea of FHE is to hide the original information through polynomials. The encryption process involves selecting a key polynomial, generating random polynomials, and a small "error" polynomial. To perform calculations, the operations need to be converted into circuit form.

The main problem faced by FHE is the accumulation of noise, which increases rapidly as the computation progresses. To address this issue, three technologies have been proposed:

• Key Switching: Compressing Ciphertext Size
• Modulus switching: controlling noise growth
• Bootstrap: Resetting noise to initial levels

Currently, mainstream FHE schemes use bootstrapping techniques to achieve unlimited depth of computation.

Challenges of FHE

The overhead of FHE computation far exceeds that of ordinary computation. Taking AES-128 decryption as an example, the computation time of the FHE version is about 500 million times that of the ordinary version. To promote the development of FHE technology, the US DARPA launched the Dprive program, aiming to increase the speed of FHE computation to 1/10 of that of ordinary computation. The program primarily focuses on the following aspects:

• Increase processor word length
• Build dedicated ASIC processors
• Develop MIMD parallel architecture

Despite slow progress, FHE technology still has unique value in protecting sensitive data, especially in the context of the post-quantum era.

Applications of FHE in Blockchain

In the blockchain field, FHE is mainly used to protect data privacy, including on-chain privacy, AI training data privacy, voting privacy, and more. Some projects view FHE as a potential solution to the MEV problem. However, fully encrypted transactions may also have negative impacts, such as reducing network throughput.

Major FHE Projects

Currently, most FHE projects use technology provided by Zama, such as Fhenix, Privasea, Inco Network, and Mind Network. These projects differ in their business models:

• Fhenix: Building a Privacy-First Optimism Layer 2
• Privasea: Focused on LLM data computation
• Inco Network: Building a Layer 1 Network
• Mind Network: Combining the Restaking Track

It is worth mentioning the Octra project, which employs innovative FHE technology based on hypergraphs.

Future Outlook

FHE technology is still in its early stages, and its development lags behind zero-knowledge proof technology. The main challenges include high costs, engineering difficulties, and unclear commercialization prospects. However, with more funding and attention, it is expected that more FHE projects will emerge.

The implementation of FHE chips is key to the commercialization of this technology. Multiple companies such as Intel, Chain Reaction, and Optalysys are exploring this field. Despite facing technical barriers, FHE, as a promising technology with clear demand, is expected to bring profound changes to industries such as defense, finance, and healthcare.