Meta Banned from Using Brazilian Data for AI Training

TapTechNews July 3rd news, according to the Associated Press, on Tuesday (2nd) local time, the Brazilian National Data Protection Agency determined that Meta, the social network giant that owns platforms such as Facebook, Instagram, and WhatsApp, shall not use data from Brazil to train its generative AI models.

Meta recently updated its privacy policy, allowing the company to use people's public posts for training its models. According to the communiqué of the Brazilian National Data Protection Agency, this move by Meta will cause severe, irreparable or difficult-to-repair damage to the basic rights of the affected data subjects, and the imposed ban is a preventive measure.

Brazil is one of Meta's largest markets. Among the country's total population of 203 million (TapTechNews note: data from the 2022 Brazilian census), just on the Facebook platform alone, there are about 102 million active users in Brazil.

A spokesperson for Meta expressed disappointment with the ban and insisted that the company's measures comply with local laws and regulations. This is a setback for the development, innovation and competition of AI, and further delays the benefits that AI brings to the Brazilian people.

In Europe, Meta's updated privacy policy has also hit a snag - the company recently suspended the plan originally scheduled to start last week, which is to use users' public content to train its own AI model. However, in the United States, due to the lack of a national law to protect online privacy, Meta's use of user data to train AI is already underway.