Meta announced on Monday that it will begin using publicly available content from European users to train its artificial intelligence models, resuming work that had been paused last year due to concerns raised by activists regarding data privacy. The company, which owns Facebook and Instagram, stated that it will use public posts and comments shared by adult users in the European Union to enhance its AI systems.
Meta also mentioned that interactions with its AI, such as user queries and questions, will be used to improve its models. This decision follows the launch of Meta’s AI assistant for European users last month, a long-awaited rollout after it had already launched in the United States and other major markets.
The company’s AI training had been delayed due to strict data privacy laws in the EU, which grant users control over how their personal information is used. Privacy group NOYB, led by activist Max Schrems, had raised concerns about Meta’s plans, filing complaints with national privacy regulators to stop the company from proceeding. However, Meta noted that a panel of EU privacy regulators had confirmed in December that its approach met legal requirements.
Meta emphasized that it would not use private messages to train its AI models and argued that its approach is in line with industry practices, citing that competitors like Google and OpenAI have already used data from European users for training their AI systems.
The company also stated that it would notify EU users about the training and provide them with a form to object if they wish. “We’ll honor all objection forms,” Meta assured.