On Monday, Meta announced that it will restart its plan to train artificial intelligence (AI) models using public content in the European Union (EU). This decision comes after the company had previously paused the initiative due to concerns from regulators about data privacy. Meta will now use public posts and comments from platforms like Facebook and Instagram to enhance its AI models. The decision was made following extensive discussions with EU regulators. Meta has assured that it will implement various safeguards to protect user privacy during the training process. These measures include rigorous data source screening to ensure that only content explicitly published by users as public is used, as well as bolstering data protection and encryption technologies. Meta emphasized that the primary goal of training AI models is to improve the quality of its services. This includes refining content recommendation algorithms, enhancing platform security, and reducing the spread of harmful information. To increase transparency, the company plans to publish more detailed reports on the AI training process, providing the public with a clearer understanding of how their data is being utilized. Despite these safeguards, some data security experts remain skeptical. They argue that even with public content, the lack of transparent practices could still raise concerns among users. Meta has acknowledged these concerns and stated its commitment to continued collaboration with regulators and external experts to refine its privacy policies and data usage guidelines. This move marks a significant step in Meta's AI development in Europe and is expected to help the company better compete in the market, enhancing its position in the EU.