Meta to use European user data for AI training after regulatory pause
Meta announced Monday it will begin training its artificial intelligence (AI) models using public content from users in the European Union, marking a significant policy shift following a year of regulatory scrutiny over data privacy concerns.
The company said it would start using public posts and comments from Facebook and Instagram users in the EU this week. Interactions with Meta AI, the company's generative AI assistant, will also be used to help train its models.
The move follows the limited EU launch of Meta AI last month, which came well after its debut in the United States and other global markets. While Meta has long used user-generated content from the U.S. to train its AI systems, EU regulators had previously blocked similar efforts due to strict privacy regulations, particularly the General Data Protection Regulation (GDPR), which mandates a legal basis for processing personal data.
In June 2024, Meta paused its plans to begin training its AI systems on EU user data after pushback from the Irish Data Protection Commission (DPC), which regulates Meta in the bloc. The DPC had been acting on behalf of several European data protection authorities.
However, in September 2024, the company said it would resume its efforts in the United Kingdom, using public posts from its UK user base. Now, the company is extending that approach to the wider EU.
“Last year, we delayed training our large language models using public content while regulators clarified legal requirements,” the company said in a blog post. “We welcome the opinion provided by the EDPB in December, which affirmed that our original approach met our legal obligations. Since then, we have engaged constructively with the IDPC and look forward to continuing to bring the full benefits of generative AI to people in Europe.”
Starting this week, EU users will begin receiving in-app and email notifications outlining Meta’s plans to use public data and interactions with Meta AI. These messages will include a link to a form allowing users to opt out of having their data used for training. Meta said it will honour all existing objection forms and any new submissions.
The company also stated that it will not use private messages or data from users under the age of 18 in the EU to train its AI systems.
“We believe we have a responsibility to build AI that’s not just available to Europeans, but is actually built for them,” Meta said. “That’s why it’s so important for our generative AI models to be trained on a variety of data so they can understand the incredible and diverse nuances and complexities that make up European communities. That means everything from dialects and colloquialisms, to hyper-local knowledge and the distinct ways different countries use humour and sarcasm on our products.”
Meta also noted that other AI developers, including Google and OpenAI, have already used European data to train their own generative models.
Meanwhile, the DPC continues to scrutinise AI training practices. Last week, the regulator announced it was investigating xAI’s training of its Grok model.
News