Today, we’re announcing our plans to train AI at Meta using public content —like public posts and comments— shared by adults on our products in the EU. People’s interactions with Meta AI – like questions and queries – will also be used to train and improve our models. This training, which follows the successful launch of Meta AI in Europe last month, will better support millions of people and businesses in the EU by teaching AI at Meta to better understand and reflect their cultures, languages and history.

What we’re doing and what you’ll see

Beginning this week, people based in the EU who use Meta’s platforms will start receiving notifications—in-app and via email—to explain the kind of data we’re going to start using, how this will improve AI at Meta and the overall user experience. These notifications will also include a link to a form where people can object to their data being used in this way at any time. We have made this objection form easy to find, read, and use, and we’ll honor all objection forms we have already received, as well as newly submitted ones.

As we’ve previously mentioned, we do not use people’s private messages with friends and family to train our generative AI models. Additionally, public data from the accounts of people in the EU under the age of 18 is not being used for training purposes.

Why we’re doing this

Last month, we launched Meta AI in the EU, which was the first step towards making Meta AI’s chat function available for free across Europe within the messaging apps people already know and love  – Facebook, Instagram, WhatsApp and Messenger.

We believe we have a responsibility to build AI that’s not just available to Europeans, but is actually built for them. That’s why it’s so important for our generative AI models to be trained on a variety of data so they can understand the incredible and diverse nuances and complexities that make up European communities. That means everything from dialects and colloquialisms, to hyper-local knowledge and the distinct ways different countries use humor and sarcasm on our products. This is particularly important as AI models become more advanced with multi-modal functionality, which spans text, voice, video, and imagery. 

It’s important to note that the kind of AI training we’re doing is not unique to Meta, nor will it be unique to Europe. This is how we have been training our generative AI models for other regions since launch. We’re following the example set by others including Google and OpenAI, both of which have already used data from European users to train their AI models. We’re proud that our approach is more transparent than many of our industry counterparts.

Our approach complies with European laws and regulations

Last year, we delayed training our large language models using public content while regulators clarified legal requirements. We welcome the opinion provided by the EDPB in December, which affirmed that our original approach met our legal obligations. Since then, we have engaged constructively with the IDPC and look forward to continuing to bring the full benefits of generative AI to people in Europe.