Google Open-Sources A First-Of-Its-Kind, General-Purpose Transpiler For Fully Homomorphic Encryption (FHE), Enabling Developers To Compute On Encrypted Data Without Being Able To Access Any Personally Identifiable Information

Google has formulated a guide of coding utilities that allow encrypted data to be fully homomorphic encrypted (FHE). The open-source set of libraries and tools enables computational operations to be performed on encrypted data without first decrypting it, resulting in increased security and privacy. Secure multi-party computing and homomorphic encryption are well-known technologies. Rather than rewriting the foundation for the technologies, FHE focuses on improving and making them appropriate for broader deployment.

It’s the first-of-its-kind general-purpose transpiler for Fully Homomorphic Encryption (FHE), which will allow developers to perform computations on encrypted data without gaining access to personally identifying information. Developers can now build safe solutions by default, private by design, and put consumers in control. It will help developers to keep their users secure online and protect their data.

Homomorphic Encryption is a method of safe computing that eliminates the need to decode your data before processing it. Instead, homomorphic encryption allows you to process ciphertexts to ensure that the encrypted outputs will match those obtained by first decrypting, processing, and encrypting the input data. Encrypted data may be sent over the Internet to a server and processed without being interpreted using FHE. FHE may also be used to train machine learning models on sensitive data privately.

Suppose you’re developing a diabetes-related app. This software may collect sensitive data from its users. You’ll need the means to keep it private and secure while simultaneously sharing it with medical professionals to get valuable insights that might lead to significant medical breakthroughs. With Google’s FHE transpiler, you can encrypt data and share it with medical professionals, who can then analyze it without decrypting it, offering helpful information to the medical community while guaranteeing that no one has access to the data’s underlying information.

FHE might potentially assist researchers in uncovering correlations between particular gene mutations in the next ten years by evaluating genetic information across thousands of encrypted samples and testing various hypotheses to locate the genes most strongly related to the diseases they’re researching.

“Spell checks for email, updates from wearables, medical record analysis, and, soon, things like photo filters or genomic analysis” are some of the applications for homomorphic encryption.

To utilize this technology, no prior knowledge of cryptography is necessary. The technology aims to address a shortage of crypto competence among developers, which has historically hampered the broader use of such tools.

Let’s take a deeper look at it.

There are two critical components to the transpiler. One is, it makes use of Google’s open-source XLS SDK to take advantage of its compilation process and transform higher-level language operations into lower-level boolean operations, as FHE requires. Secondly, converting from the intermediate form supplied by XLS to an HFE calculation employs Google’s TFHE fully homomorphic encryption library.

According to Google, this modular architecture provides several benefits. First, due to XLS, a variety of high-level languages are supported out of the box. Similarly, the FHE-ready code may be written in any language with an FHE library with logical gates as part of its API.

There’s a significant catch. XLS does not support all C++ features. Variable-length arrays, while-loops and for-loops with a variable termination condition, and floating-point data are among the features that aren’t supported. Furthermore, both XLS and TFHE are still in the early stages of development and are likely to undergo considerable changes.