Table of Contents
According to the popular Mozilla company, Google Floc is a flop and unsecured premise privacy sandbox that does not protect the privacy and could lead to the serious risks.
What is Floc Originally?
FLoC is premised on a compelling idea: enable ad targeting without exposing users to risk, says Eric Rescorla, author of the TLS standard and the chief technology officer of Mozilla.
Several privacy properties in the current design could result in significant risks if widely implemented in its current form.
FLoC stands for Federated Learning of Cohorts, a initiative that seeks to develop alternative solutions to satisfy cross-site use cases without resorting to third-party cookies or opaque tracking mechanisms.
It enables marketers to guess users’ interests without requiring individual identification, eliminating the privacy implications of targeted advertising. Currently, targeted advertising relies on techniques such as tracking cookies and device fingerprinting that reveal users’ browsing history across websites to advertisers and ad networks.
What Else Does It Do?
FLoC avoids the cookie by using a new “cohort” identifier that buckets users by their similar browsing behaviors.
This information can be aggregated by advertisers to build a list of websites that users in a cohort visit, rather than using the visits made by the specific user, and then advertisers can target ads based on the cohorts’ interests.
What Does Mozilla Have To Say about Floc?
Mozilla said that FLoC profiles can provide additional information about the FLoC as a whole. Taking into account the individual profiles, we can make generalizations about the FLoC cohort as a whole.”
In addition, the cohort IDs assigned to users are recalculated on the device weekly, which is intended to reflect their evolving interests over time and prevent them from being used as persistent identifiers to track users.
At some point next year, Google plans to replace third-party cookies with FLoC in its Chrome browser as an origin trial.
This conclusion is supported by Mozilla’s FLoC analysis. Even though a few thousand users share a cohort ID, it is possible to narrow down the list of users very quickly using fingerprinting data and the periodically recalculated cohort IDs as an easy way to differentiate users from one to the next.