Fhenix to bring confidential computing to public blockchains

Fhenix, a blockchain research and development company, is positioning itself as a full-stack infrastructure for confidential decentralized finance (DeFi) designed to bring encrypted computation directly onto public blockchains.

Fully Homomorphic Encryption (FHE) is at the center of the new strategy, according to a press release shared with Finbold on February 17 and a livestream on X the day prior featuring Fhenix founder Guy Zyskind.

This cryptographic breakthrough allows computations to be performed while data remains encrypted. Such an approach, the management argues, eliminates exposure at every stage from execution to settlement, which puts FHE ahead of Zero-Knowledge proofs, Trusted Execution Environments, and Multi-Party Computation.

“Privacy is, I believe, the most difficult problem to solve in blockchains… Many of them, even if you look at ZK projects, many of them went to scalability because that’s actually the easier problem to solve. Adding privacy on top of ZK, something like what Fhenix is doing, trying to build and scale Fully Homomorphic Encryption? Those are very, very hard problems to solve. Very few people can do that,” Zyskind said on X.

Enhanced blockchain privacy

Among the most notable innovations was CoFHE, an FHE coprocessor designed to offload encrypted computation from the main chain. Recently deployed on Base, this stateless engine is intended to make private smart contracts viable at scale, delivering throughput improvements of up to 5,000 times over earlier FHE systems.

Another key feature, fhEVM, allows developers to write privacy-preserving applications using familiar Solidity tooling. Instead of learning an entirely new stack, developers can thus integrate encrypted execution into an existing Ethereum (ETH) compatible environment.

On the encrypted verification front, renewed focus on DBFV signals continued work on making encrypted computations not only private but verifiable in decentralized environments. Together, these functionalities are meant to combat issues such as data leaks, which are plaguing AI agents.

“AI agents: they are really bad at security and privacy right now. They leak all information…. We need to find ways to protect data and protect it, basically, like Fort Knox,” Zyskind added.

Programmable digital privacy and institutional security

With features like Shielded Mode for end-to-end encrypted payments and experiments integrating privacy into HTTP 402 payment standards dubbed Fhenix402, the company believes the recent innovations have numerous implications for broader Web3.

The information shared with Finbold cited private governance votes, encrypted identities, confidential business analytics, and front-running protection as some examples. More notably, however, the livestream on X referenced institutional interest.

As discussed by the speakers, J.P. Morgan had already approached Fhenix in the past, exploring the tokenization of what was described as $1.5 trillion in assets under management. Unfortunately, such efforts were impossible without customer-level privacy, but the example shows how large financial players already recognize privacy as a structural requirement, not a luxury.

Featured image via Shutterstock