Fhenix, a blockchain research and development company, is establishing itself as a comprehensive infrastructure for confidential decentralized finance (DeFi) aimed at integrating encrypted computation directly onto public blockchains.
The core of their new strategy lies in Fully Homomorphic Encryption (FHE), as highlighted in a recent press release shared with Finbold on February 17 and a livestream on X featuring Fhenix founder Guy Zyskind.
This groundbreaking cryptographic advancement enables computations to be carried out while data remains encrypted. This approach, according to the management, eliminates exposure at every stage from execution to settlement, setting FHE apart from 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.
https://t.co/XgmbxOf9bb
— Fhenix (@fhenix) February 16, 2026
Enhanced blockchain privacy
One of the notable innovations is CoFHE, an FHE coprocessor designed to offload encrypted computation from the main chain. Recently deployed on Base, this stateless engine aims to make private smart contracts viable at scale, achieving throughput improvements of up to 5,000 times over previous FHE systems.
Another significant feature, fhEVM, enables developers to create privacy-preserving applications using familiar Solidity tooling. Instead of learning a completely new stack, developers can seamlessly integrate encrypted execution into an existing Ethereum (ETH) compatible environment.
In the realm of encrypted verification, a renewed focus on DBFV signals ongoing efforts to make encrypted computations not only private but also verifiable in decentralized environments. These functionalities aim to address issues such as data leaks, which are a concern for 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 known as Fhenix402, the company envisions various implications for the broader Web3 ecosystem.
The information provided to Finbold mentioned private governance votes, encrypted identities, confidential business analytics, and front-running protection as examples. Moreover, the livestream on X highlighted institutional interest.
As discussed by the speakers, J.P. Morgan had previously engaged with Fhenix, exploring the tokenization of assets totaling $1.5 trillion under management. However, such endeavors were hindered by the lack of customer-level privacy. This example underscores how major financial institutions recognize privacy as a fundamental requirement, not a luxury.
Featured image via Shutterstock
