1
头图

Google announced that has open sourced the first Fully Homomorphic Encryption (FHE) universal translator under the Apache License 2.0. The source code hosted on GitHub.

image.png

With FHE, encrypted data can traverse the Internet to a server, where it can be processed without being decrypted. Google's transcoder will enable developers to write code for any type of basic calculation, such as simple string processing or mathematics, and run it on encrypted data. The transcoder will convert these codes into a version that can be run on encrypted data. In this way, developers can create new programming applications that do not require unencrypted data. FHE can also be used to train machine learning models on sensitive data in a private way.

For example, imagine you are building an application for people with diabetes. This application may collect sensitive information from users, and you need a way to keep this data private and protected, while also sharing this data with medical experts to understand valuable insights and thereby achieve important medical advancements. Through the transponder provided by Google for FHE, you can encrypt the data you collect and share it with medical experts, who can analyze the data without decrypting it-to provide useful information for the medical community, At the same time, ensure that no one can obtain the basic information of the data.

In the next 10 years, FHE can even help researchers find the association between specific gene mutations by analyzing the genetic information of thousands of encrypted samples and testing different hypotheses to determine the genes most closely related to the diseases they are studying. . Privacy-protecting technologies are at the forefront of Google's innovation, and they have shown great potential to help shape a more private Internet.

In 2016, Google researchers invented Federated Learning, a technology that helps protect privacy by keeping as much personal information as possible on your device. In 2019, Google provided its differentiated privacy library to any organization or developer for free, which is an advanced anonymity technology that enables developers to learn from their data privately.

Google conveniently said: "Opening our technology to the developer community for feedback and use will help make them better. We will continue to invest in and lead in the field of privacy protection technology, publish new work, and open source it for everyone to use on a large scale-we are happy to continue this practice by sharing this latest development with developers around the world. We can’t wait to see what you will build, and we look forward to working together on the journey to a safer Internet. 』

Related Links:

  1. https://github.com/google/fully-homomorphic-encryption
  2. https://developers.googleblog.com/2021/06/our-latest-updates-on-fully-homomorphic-encryption.html

鸣飞
1.7k 声望42 粉丝

SF 思否社区编辑一枚,关注人工智能、云计算、开源和行业的技术动态,欢迎投喂优质资讯!