Menlo Park, CA – Meta has announced a major update to its data usage policy in the European Union, revealing that beginning May 27, 2025, it will start using certain types of personal content from Facebook and Instagram users to train its artificial intelligence models. The policy shift marks one of the most significant efforts by a major tech company to integrate user-generated content into large-scale AI development under the General Data Protection Regulation (GDPR) framework.

Under the new policy, Meta will by default include photos, posts, captions, and comments shared by users in its AI training datasets. However, the company confirmed that users will have the ability to opt out of this data usage via a dedicated request form accessible through its Help Center and privacy settings. The opt-out process is being introduced to meet GDPR requirements for transparency, user consent, and the right to object to data processing.

Meta’s decision comes in response to evolving legal interpretations within the EU, which increasingly recognize the possibility of using personal data for AI model training—as long as individuals are properly informed and provided with meaningful control over their data. The company emphasized that only content already shared with Meta, either publicly or within the scope of existing user permissions, will be considered. Private messages, for example, will not be included.

According to Meta, the personal data collected will be anonymized and aggregated before being processed by its AI systems. The company states that the initiative aims to enhance the quality and safety of AI-powered features across its ecosystem, including Facebook, Instagram, and WhatsApp, and to deliver more personalized, secure, and contextually aware user experiences.

“We’re committed to building AI responsibly and transparently,” a Meta spokesperson said. “This update will help improve tools like content recommendations, language translation, content moderation, and even user safety features—while ensuring that individuals retain control over how their data is used.”

Still, the announcement is likely to spark ongoing debate over how much user data should be used in AI training, particularly as governments and regulators scrutinize how tech giants collect, store, and repurpose personal information. Privacy advocates have already raised concerns that even with opt-out options, many users may remain unaware of the implications of default data use in such training.

To address these concerns, Meta stated it is investing in user education campaigns to inform individuals across the EU about their rights under the GDPR and how to manage their data preferences. The company also pledged to periodically review and update its AI training policies to reflect evolving legal standards and ethical guidelines in the rapidly changing AI landscape.

As the use of AI accelerates across industries, Meta’s approach could set a precedent for how other platforms leverage user content while navigating Europe’s stringent privacy regulations.