BRUSSELS, July 12 — The European Commission’s formal proceeding against Meta, opened in April 2024 under the Digital Services Act, examines how the company uses user data to train its AI models. The technical reality is that every scrap of text, image, and interaction a user generates on Facebook or Instagram is potential fodder for the next generation of large language models and image generators. Under the hood, Meta’s AI ambitions require vast datasets, and the company has been pulling from its European user base without, regulators argue, adequate consent.
The General Data Protection Regulation, the EU’s flagship privacy law, imposes fines of up to 4% of global annual revenue for violations. That is not a theoretical ceiling.
In January 2023, the Irish Data Protection Commission fined Meta €390 million ($425 million) for GDPR violations tied to its advertising practices. The detail that matters: Ireland is Meta’s lead regulator in Europe because the company’s EU headquarters are in Dublin, and the Irish DPC has been the primary enforcer of GDPR against the big US tech platforms.
The American front
Across the Atlantic, Meta faces a separate but parallel regulatory squeeze. The Federal Trade Commission is investigating Meta’s AI practices as part of a broader antitrust case filed in December 2020. That case seeks to unwind Meta’s acquisitions of Instagram (2012) and WhatsApp (2014), arguing the company bought its way to dominance rather than competing on merit.
The FTC’s interest in AI is newer, but it fits a pattern: regulators on both sides of the Atlantic are asking whether Meta’s control of user data gives it an unfair advantage in the AI race. A class-action lawsuit filed in California in November 2024 adds another layer.
The plaintiffs allege that Meta’s AI training data includes copyrighted content without permission. This is the same legal theory that has roiled the generative AI industry since the launch of ChatGPT: companies scrape the open web, absorb copyrighted books, articles, images, and code, and then profit from models that can reproduce that material. Meta is not alone in the crosshairs, but it is a particularly large target because of the sheer volume of data it holds.
Investors look the other way
Despite the regulatory headwinds, Meta’s stock has gained. The market seems to be betting that AI growth prospects outweigh legal risks. As ever, the detail that matters is the gap between regulatory process and market reality.
Investigations take years. Fines, even large ones, are a cost of doing business.
The real threat — forced divestiture of Instagram and WhatsApp, or a ruling that Meta must delete its AI training data — is distant and uncertain. Investors are pricing in the near-term revenue opportunity from AI-powered features, not the theoretical worst case. The European Commission has not set a deadline for its DSA investigation.
The FTC case is moving through the US court system, with no trial date yet set. The California class action is in early stages.
For now, Meta continues to train its models, and the data keeps flowing.
Sources
{“@context”: “https://schema.org”, “@type”: “NewsArticle”, “headline”: “Meta Faces EU, US Scrutiny Over AI Data and Competition”, “description”: “BRUSSELS, July 12 — The European Commission’s formal proceeding against Meta, opened in April 2024 under the Digital Services Act, examines how the company uses…”, “datePublished”: “2026-07-12T09:30:00+00:00”, “author”: {“@type”: “Person”, “name”: “Sebastian Smith”}, “publisher”: {“@type”: “Organization”, “name”: “InfoPulse Today”, “url”: “https://newsfeed.2bme.xyz/”}, “citation”: [{“@type”: “CreativeWork”, “url”: “https://en.wikipedia.org/wiki/General_Data_Protection_Regulation”}]}
