Meta Platforms feeds its AI with data from European customers

Meta Platforms is resuming the training of its AI models with data from the EU. Anyone who does not wish to donate their data must take action.

listen Print view
Meta logo and head of a female robot

(Image: Below the Sky/Shutterstock.com)

2 min. read

Meta Platforms is once again using personal data from EU customers to train its artificial intelligence (AI). The American data company announced this on Monday. This affects profiles, posts, comments, and other public statements made by Whatsapp, Threads, Instagram and Facebook users in the European Union, as well as unpublished interactions with Meta's language models there.

“Making AI Work Harder for Europeans”, that's what Meta calls it. Anyone who does not want their data to be harvested in this way must opt out in their account settings. Meta promises not to touch private messages between users or public posts from accounts whose owners are registered as under 18 years of age.

Originally, the company had already set out to access the treasure trove of data last year, but was stopped by the Irish data protection authority responsible for Meta. In December, the data protection authorities of the EU member states then agreed on a common approach. According to this, data companies can in principle invoke “legitimate interest” as the legal basis for the processing of personal data by AI models.

The European Data Protection Board (EDPB) has come up with a three-stage test for this. The first step is to check whether the claim to data processing is legitimate. This is followed by a “necessity test”, which must show that the data processing is necessary. Finally, the fundamental rights of the data subjects must be weighed up against the interests of the respective AI provider.

Videos by heise

Meta Platforms thinks that its interests outweigh the fundamental rights of the data subjects and speaks of “constructive” interaction with the Irish data protection authority. Meanwhile, the same authority suspects that Elon Musk's company xAI does not meet the requirements. This company uses personal data of EU users from public contributions to its subsidiary X to train its AI model Grok. The Irish authorities have launched an investigation.

(ds)

Don't miss any news – follow us on Facebook, LinkedIn or Mastodon.

This article was originally published in German. It was translated with technical assistance and editorially reviewed before publication.