Meta AI Expands in EU with Localized Training Data
The training data will focus on understanding and reflecting European cultures, languages, and history, with the goal of creating AI models that are more effective and tailored to local needs. This move is seen as a positive step towards more transparent and responsible AI development, and is expected to have a significant impact on the future of AI in the EU.
Meta will begin training its AI models on public content and interactions from European users to better support the region's diverse cultures, languages, and history, while providing users with the option to opt out of data usage for training purposes.
The move is part of Meta's efforts to launch its AI technology in Europe, following a delayed introduction due to regulatory scrutiny. The company plans to use public content from EU users, excluding private messages and data from users under 18, to improve its AI understanding of European dialects, cultural nuances, and humor. This approach is seen as more transparent than many of its industry counterparts, with Meta set to feature its main AI model, Llama, at the upcoming LlamaCon event.
Meta will notify EU users about the types of data being used for AI training, including interactions with Meta AI and public content shared by adults. Users will also have the option to object to their public data being used for AI training, providing an additional layer of transparency and control. The company believes it has a responsibility to build AI that is tailored to European needs and complies with European laws and regulations.
As Meta continues to expand its AI capabilities in the EU, the company's commitment to transparency and user control is expected to be closely watched by regulators and users alike. With a strong focus on localized training data, Meta aims to provide more effective support for people and businesses in the region, while also addressing concerns over data privacy and transparency.