Brazil's data protection authority (ANPD) has prohibited Meta from using Brazilian personal data to train its artificial intelligence (AI) models, citing potential serious harm and user difficulties. This decision comes after Meta updated its privacy policy in May to allow the use of public data from Facebook, Messenger, and Instagram in Brazil for AI training. The ANPD's move follows a report by Human Rights Watch, which found that a significant image-caption dataset used for AI training contains identifiable photos of Brazilian children, posing risks of exploitation. With Brazil being one of Meta's largest markets, the ANPD has given Meta five days to comply with the order or face daily fines. Meta claims its updated policy complies with Brazilian privacy laws and expresses disappointment with the ruling, stating it hinders innovation and AI development. The company also argues that users can opt out of data usage for AI training, but the ANPD contends that there are unjustified obstacles to doing so. In a related development, Meta's AI assistant will not be available in the European Union (EU) due to privacy concerns raised by Ireland's Data Protection Commission. The company had been collecting publicly shared content from Facebook and Instagram users worldwide to train its large language model (LLM), but the EU's stringent data protection regulations, including the General Data Protection Regulation (GDPR) and the EU AI Act, have led to legal challenges and a delay in launching the AI assistant in Europe. Meta's decision to pause its plans to use social media posts for AI training has faced objections from European advocacy groups, and users in the EU and the UK have the right to object to their data being used for this purpose. When the delayed plans are reinstated, users will be notified and given the opportunity to object before the changes take effect. Overall, the ANPD's ban on Meta's use of Brazilian personal data for AI training, along with the EU's regulatory challenges and user objections, reflects the growing global scrutiny and regulation of AI training practices, particularly concerning privacy and data protection.
