April 2024 Summaries

Zama's fhEVM introduces confidential smart contracts to blockchain technology, allowing for new applications in areas like DeFi and gaming by addressing the challenge of public on-chain data. As the first private smart contract protocol utilizing fully homomorphic encryption (FHE), fhEVM ensures that transaction data and on-chain states remain encrypted throughout the entire process, even during execution. The tutorial by Zama team member Joseph-Andre Turk demonstrates how to implement asynchronous decryption in Solidity contracts using fhEVM. Additionally, users are encouraged to engage with Zama's fhEVM GitHub repository, consult its documentation, participate in community channels for support, and contribute to advancing the FHE space through Zama's Bounty and Grant Programs.

Zama has released new versions of its homomorphic encryption products, including TFHE-rs v0.6, Concrete v2.6, Concrete ML v1.5, and fhEVM v0.4, each offering significant advancements to enhance accessibility, ease, and speed. TFHE-rs v0.6 introduces Zero-Knowledge Proofs, enhanced GPU support, and additional cryptographic features, while Concrete v2.6 brings improvements in programmable bootstrapping and function composition. Concrete ML v1.5 offers a new DataFrame API, increased neural network speed, and improved FHE simulation mode. The updated fhEVM v0.4 includes new encrypted types, an asynchronous decryption API, and a standard contracts library. Alongside these updates, Zama has revamped its developer resources and documentation to provide a more intuitive coding experience, with plans to further enhance content readability in the near future.

TFHE-rs v0.6 introduces significant updates, notably enhancing cryptographic capabilities by incorporating Zero-Knowledge Proofs to verify the correctness of public key encryption without revealing the encrypted message. This update also improves GPU support, particularly for signed integer operations and parallel processing, allowing faster computations compared to CPUs. The release includes a new C API enabling cross-language support and provides performance improvements such as stabilized multi-bit Programmable Bootstrapping (PBS). The advanced use of H100 GPUs, which offer better performance over previous GPUs like V100, is highlighted, showcasing their capability in large-scale machine learning tasks. Additional enhancements involve operations like checked division, improved multiplication speed, and new noise distributions, with future plans to reduce ciphertext sizes and support multi-GPU computations, all aimed at bolstering efficiency and performance in cryptographic applications.

Concrete ML v1.5 introduces several enhancements, including a new DataFrame API that allows encrypted data manipulation, advancing its primary use case of private inference and confidential collaboration. This release also offers a feature that accelerates neural network inference by 2-3 times through a trade-off between prediction speed and precision, with an improved FHE simulation mode to assess the impact on accuracy. Furthermore, a new demo showcases secure text data anonymization for querying knowledge bases with ChatGPT without exposing personally identifiable information, enabling the creation of anonymized knowledge bases for retrieval-augmented generation. The DataFrame API draws inspiration from the Pandas package, providing capabilities for encrypting, joining, and decrypting data, facilitating encrypted collaboration and model training.

The latest version of fhEVM, v0.4, introduces new encrypted data types and an asynchronous decryption API to enhance performance and efficiency. This update supports encrypted booleans, 4-bit and 64-bit integers, and addresses, enabling more cost-effective operations and expanding use cases for privacy-preserving decentralized applications. The asynchronous decryption API allows smart contracts to request decryption via an oracle and receive results through callbacks, freeing resources and optimizing scalability compared to previous synchronous methods. This feature replaces synchronous decryption, which will be deprecated in the next version. Additionally, a new standard library of contracts, including ERC20 and DAO, has been released to streamline development and foster innovation. Developers are encouraged to upgrade their code to utilize these new features, with resources available for support and further information.

Concrete v2.6 introduces several advancements to enhance the performance, efficiency, and functionality of its framework, focusing on approximate programmable bootstrapping (PBS), input compression, extended function composability, and faster simulations. Approximate PBS involves improved rounding techniques that increase the speed of non-linear operations, beneficial for applications tolerant to inexact calculations like deep learning tasks. Input compression is further optimized to significantly reduce bandwidth and disk space usage, with input ciphertexts becoming much smaller. The update also advances function composability, allowing developers to define a set of functions within modules, enabling more flexible programming over encrypted data. The simulation speed has been improved dramatically, making development more efficient without requiring changes to the existing API. Enhancements in operators such as ternary-if, Relu, and Sign functions lead to faster processing, contributing to the overall efficiency of the platform. These improvements are designed to be seamlessly integrated into current workflows, encouraging users to explore the new features without needing to modify their existing setups.