Did you know Meta Prepares to Train AI Models on EU Users’ Data Taken from Facebook and Instagram

 

Despite all the controversy that arose, tech giant Meta is now preparing to train its AI systems on data belonging to Facebook and Instagram users in the EU.

Facebook’s parent firm made it very clear how features like comments, history of chats, and posts going public would be a part of the data in use. However, any private or confidential matters like exclusive texts to loved ones or pals won’t be included. Also, the age restriction is 18 years and above for this endeavor.

As per Meta, it would begin to alert users in the EU about how training will ensue this week, and it would be done through email and in-app alerts. Anyone who doesn’t feel the need to be included can resort to the objection form and select the opt-out status.

They can find links through the platform’s privacy policy. This might be one critical factor why we’ve seen such a delay in training for Meta AI through user data. Irish data regulators questioned the act and needed Meta to halt the operations on this front.

The company shared how it is now training using data belonging to EU users to design models that portray more about how they’re being used in that part of the world. From knowledge to subjects of interest and so on, there’s a lot of information to gather. The tech giant shared how it matters a lot for things like video, text, pictures, and voice rolled out by the AI model.

The news comes after the company’s reveal in 2024 about training AI systems using data from UK users. The latter are also given a lot of protection in terms of using personal data on the web than those located in America. What the tech giant can receive now is very small compared to what it may already have when it trained using images and text from users on Facebook that went public in 2007.