In yet another disappointing development for the users asking for Privacy, Google has expressed dismay over blocking cookies by the users. Google says that blocking the cookies for browsers does not promote high efficiency of the applications and hampers performance.
Recently Googles Privacy Sandbox has been experimenting with newer technologies like blockchain and Federated Learning.
Google Says Blocking Cookies is not Good
The cookies are used by the companies to collect user data and then use it for targeted advertisements. Google says that the large scale cookie blocking derates the privacy of users by promoting opaque techniques such as fingerprints.
Fingerprinting is a technique that developers utilize to refrain the checks that are used by cookies. By bypassing the Cookie Restrictions these techniques collect user data and other pieces of information like font and device. The collected information can generate specific identifiers to co-relate users on different websites. This is again a threat to user privacy.
According to Google Spokesperson, the users cannot clear their fingerprint on the web, unlike cookies, thus bereaving them of the control over the way their information is gathered and processed.
Blocking Cookies affect AD business
Google Sandbox privacy is intended to work for and with each user and entity on the internet.
According to Google, the various privacy features that it is testing through the Sandbox will initiate a process of creating a balanced and homogenous solution for everyone. This privacy solution will be highly proficient in taking opinions and understand the needs of the various stakeholders.
Google is using Federated learning
Google is also using (and promoting to the data science community) a technology known as federated learning to determine how ads can be targeted by clustering people into groups without revealing any data to advertisers or even having any personally identifiable data leave the browser.
Recently Google is leveraging the newest technology of Federated Learning in which it consorts to find ways to create targeted ads by combining users in relevant groups and still hiding users data from the advertisers. The concept has grown over the years helping data scientists to program machine learning and artificial intelligence models based on real-time user data without the need for data leaving users computers or device.