Building a more private web
Federated Learning of Cohorts (FLoC)
Federated Learning of Cohorts (FLoC) is a new way for advertisers and sites to show relevant ads without tracking individuals across the web. FLoC was developed with the idea to protect individuals' privacy by placing them in a large crowd — a cohort — of thousands of people with similar recent browsing activity without any of them being individually identified.
![PSB FLoC Video Cover](https://www.gstatic.com/marketing-cms/assets/images/73/7e/95e0e67848b6b3db16d3a16125ed/thumnbnail-floc.png=n-w964-h543-fcrop64=1,38150000c823ffff)
How it works
The Federated Learning of Cohorts (FLoC) uses an open source technology that looks for clusters of similar information. In this case, it’s used to create cohorts of people based on their web browsing activity. For example, one cohort can be thousands of people who have recently visited websites about gardening and travel overseas; another cohort can be a group of people who have recently visited sites about art supplies, cooking, and hiking. Your browser uses the FLoC technology to determine which cohort corresponds most closely to your recent web browsing history.
Privacy improvements
You’re part of a crowd. FLoC keeps you among a large group of people with similar browsing history. This group isn’t based on who the individuals are, but rather on their collective interests so advertisers can still show relevant ads.
Your personal browsing history doesn’t leave your browser or your device, and it’s not shared with anyone, including advertisers. This means you can continue to see relevant ads and content without needing to be tracked across the web.