FHE Onchain (@FHEOnchain) header
FHE Onchain (@FHEOnchain) profile picture

Sharing the latest insights and breakthroughs on the intersection of privacy-enhancing technologies (PETs) and Web3.

Joined May 2025
Photo by FHE Onchain (@FHEOnchain)Photo by FHE Onchain (@FHEOnchain)Photo by FHE Onchain (@FHEOnchain)Photo by FHE Onchain (@FHEOnchain)Photo by FHE Onchain (@FHEOnchain)Photo by FHE Onchain (@FHEOnchain)
As interest in data privacy grows, @Apple has begun leveraging Fully Homomorphic Encryption (FHE) to secure machine learning (ML) processes without compromising user privacy. Here’s a breakdown of Apple’s approach and how FHE is applied to sensitive data handling👇
24
8
410
Additionally, Apple has open sourced its Swift Homomorphic Encryption library, potentially encouraging broader development of privacy-preserving applications that leverage server-side data without exposing sensitive information.
3
3
208
Apple’s FHE integration highlights how homomorphic encryption can be applied at scale for secure, privacy-focused ML use cases. Whether for caller ID, image recognition, or server lookups, these techniques emphasize the potential of FHE in safeguarding user data across digital services.
1
2
169
Blind auctions is one of the most compelling use cases of fhEVMs. If you want to learn more on how they work behind the scenes, then check out our latest blog post. 👇
1
111
In TFHE, encryption can be either private or public—depending on whether the sender knows the secret key. Let’s walk through how a private setup works, how we convert it to a public one, and how public key encryption works under the hood 👇
1
78
This gives an encryption of the message—without ever needing S. This is the core idea behind TFHE’s public key approach. It is also possible to perform sample extraction to convert the GLWE ciphertext into its LWE equivalent.
1
64
Depending on the use case, either the private key setup, or the public key setup may be more suitable. So this choice should be made accordingly.
61
Fully Homomorphic Encryption (FHE) became possible from early privacy homomorphisms to Gentry’s breakthrough to today’s hardware-accelerated libraries. Watch ↓
19
9
339