X faces lawsuit over AI data use in 9 EU countries

NOYB, a digital rights non-profit organisation, filed a complaint against X in nine European countries accusing the social network of using personal data to train its ‘Grok’ AI model without user consent. The Austrian-based group says that X, formerly Twitter, violates EU privacy laws.

In a blog post, the NOYB chairman, Max Schrems, announced that the rights group had filed General Data Protection Regulation (GDPR) complaints with data privacy authorities in Austria, Belgium, France, Greece, Ireland, Italy, the Netherlands, Spain, and Poland.

Rights group wants full-scale investigation into X

NOYB wants to pressure the Irish Data Protection Commission (DPC) to fully investigate allegations that X used data from 60 million users in the EU without their consent or knowledge. The DPC is the lead European Union regulator for most of the big U.S. tech companies headquartered in Ireland.

https://twitter.com/EasyBakedOven/status/1816696187765838146?t=e3-P8f_npfdArJUYb25qyg&s=19

The authority has taken court action against X to stop the processing of user data to develop, refine or train its AI. The DPC wants to enforce an order to bring the Elon Musk-owned company’s systems into compliance with the GDPR. However, NOYB accuses the Irish DPC, describing it as “notoriously pro-corporate, and taking half-hearted action against X.”

Schrems, the NOYB chairman and privacy activist, said the DPC’s court action did not question the legality of X processing user data. Instead, he argued, they were concerned with so-called ‘mitigation measures’ and a lack of cooperation by X and shied “away from the core problem.”

“We have seen countless instances of inefficient and partial enforcement by the DPC in the past years,” Schrems said. “We want to ensure that Twitter fully complies with EU law, which – at a bare minimum – requires to ask users for consent in this case.”

NOYB is aiming to achieve a full-scale investigation into how X managed to train its generative artificial intelligence model, Grok, without consulting its users.

“Companies that interact directly with users simply need to show them a yes/no prompt before using their data,” Schrems said. “They do this regularly for lots of other things, so it would be possible for AI training as well.”

X launched Grok in November 2023, and this legal action comes a few days before the release of the beta version of Grok 2.

Major tech firms have struggled with navigating ethical issues surrounding data training. In June 2024, Meta was hit with a lawsuit in 11 European countries over its new privacy policy, which showed the company’s intention to use data generated by accounts to train its AI models.