Google open sources its differential privacy framework

Google has today announced the launch of an open-source version of the differential privacy framework that powers many of the company’s data analysis systems.

Differentially-private data analysis is an approach that enables organizations to learn from the majority of their data while simultaneously ensuring that those results don't allow any individual's data to be distinguished or re-identified.

"Today, we're rolling out the open-source version of the differential privacy library that helps power some of Google’s core products," says Miguel Guevara, product manager of Google's Privacy and Data Protection Office writing on the company's blog.  "To make the library easy for developers to use, we’re focusing on features that can be particularly difficult to execute from scratch, like automatically calculating bounds on user contributions. It is now freely available to any organization or developer that wants to use it."

Google is open-sourcing the tools that have enabled it to build this technique into features like 'Wait Times' in Google Maps where it’s been using differential privacy since 2014. Making it publicly available for free means developers, NGOs, and other public sector entities will be able to perform innovative data analysis while analysing less data to begin with.

To make the library easy for developers to use, it's focused on features that can be particularly difficult to execute from scratch like automatically calculating bounds on user contributions. It will be freely available to any organization or developer that wants to use it.

Image credit: Willy Barton / Shutterstock