Meta played with fire by wanting to train an AI with data from European users: regulators have just prevented it

Meta for now will not use the public content of the Facebook and Instagram users of the European Union (EU) and the United Kingdom to train artificial intelligence (AI) models. This is a temporary measure that emerges from the American technology company in response to pressure from several European regulators.

Ad It comes days after Ireland’s Data Protection Commission (DPC) request suspension of Meta’s controversial initiative on behalf of several data protection authorities in the bloc. The UK Information Commissioner’s Office (ICO) concern had also been expressed.

Meta has been forced to change her plan

Those led by Mark Zuckerberg have made it clear in a statement published this Friday that They are “disappointed” with the request of the DPC. In this sense, they have also pointed out that the current scenario is “a step backwards for European innovation and competition in the development of AI.”

The social media giant was scheduled to begin using public status updates, images, comments and interactions from Facebook and Instagram users starting June 26 to train AI models. As we can see, this is now suspended until further notice.

It should be noted that Meta is convinced that the aforementioned activity complies with current regulations, specifically with the General Data Protection Regulation (GDPR). According to the company, they are based on “legitimate interests” that other industry players have also used to train their AI models.

Meta says it needs data from European users so its AI models can reflect “the diverse languages, geography and cultural references of people in Europe.” For the moment we have to wait to find out what the regulators’ next move will be in this battle for user data.

In any case, if the American company is victorious in the EU and manages to avoid the obstacles it has encountered along the way, users should still have an alternative: make a series of adjustments to your profiles to prevent your data from being used to train artificial intelligence models.

Images | instagram | Christian Lue | Solen Feyissa

In Xataka | Big Tech is realizing something: its AIs can’t screw up so much