Export
The link always opens the latest version of this page.
Ask an AI about it
Open in ChatGPTA new chat that reads this entry from the site Open in ClaudeThe same, in ClaudeShare the link instead of a file, and they will always see the latest version.
Files are a snapshot of today. The live page has anything newer.
HEIR
Open-source compiler toolchain for converting AI models to operate on encrypted data using homomorphic encryption
HEIR is an open-source 1 compiler toolchain designed to convert artificial intelligence models to operate on encrypted data using homomorphic-encryption.
The toolchain enables the execution of AI inference on encrypted inputs without requiring decryption, preserving data privacy throughout the computation process. By leveraging homomorphic encryption capabilities, HEIR allows models to perform computations directly on ciphertext while maintaining the security properties of the encrypted data.
As a compiler framework, HEIR abstracts away low-level homomorphic encryption implementation details, allowing developers to work with standard AI model formats and automatically generate encrypted-data-compatible code. This bridges the gap between conventional machine learning workflows and privacy-preserving computation requirements.
Facts 3
| Attribute | Value | Confidence | Sources | Cited |
|---|---|---|---|---|
| Category | compiler | low | 1 | 1 |
| Type | dual-use | low | 1 | 2 |
| Open source | yes | low | 1 | 1 |
Developed by 1
Sources 2
- Google is making private AI practical with homomorphic encryption · Hacker News (YC)
- Updates on HEIR, the Homomorphic Encryption Compiler Project · Hacker News (YC)