Meta Platforms announced on Monday that it will start using adult user interactions and public posts from its platforms in the European Union to train its artificial intelligence models

The decision follows the delayed European launch of Meta’s AI technology, initially announced in June 2024. While the AI features were introduced in the United States in 2023, the roll-out in Europe was postponed due to regulatory scrutiny, particularly around data privacy and transparency.

Meta, the parent company of Facebook and Instagram, confirmed that users in the EU will begin receiving notifications detailing the types of data being used for AI training. These will include interactions with Meta AI such as questions and prompts, as well as public content shared by adults.

The company stated that private messages and data from users under 18 will not be included in the training process. A dedicated form will also be available for EU users wishing to object to their public data being used.

The European Commission declined to comment on Meta’s recent decision.

Meta had previously postponed the European launch of its AI models in June, following a request from Ireland’s Data Protection Commission (DPC) to delay plans involving data from social media posts.

The move also drew criticism from the privacy rights group NOYB, which called on national regulators to block the use of such content.

Ireland’s DPC is also examining other tech firms, including Elon Musk’s X and Alphabet’s Google. X is under investigation for using personal data from EU users to train its AI tool, Grok.

Meanwhile, the regulator launched an inquiry into Google last September to assess whether it safeguarded user data before deploying it for AI development.