Meta Platforms has announced that it will begin training its AI models using public data shared by adults across its platforms in the European Union. This includes posts, comments, and interactions with Meta AI on Facebook, Instagram, WhatsApp, and Messenger.
This comes after Meta paused its AI training initiatives in the region last year due to regulatory concerns. Following clearance from the European Data Protection Board (EDPB) and ongoing cooperation with the Irish Data Protection Commission (IDPC), Meta is now reactivating its efforts, this time with more transparency and user control.
What Kind of Data Will Meta Use for AI Training?
The scope of Meta’s training data includes:
- Public content shared by adults (posts, comments)
- User interactions with Meta AI features, such as AI stickers, image prompts, and conversations with the chatbot
Importantly, private messages between users and content from users under 18 will not be used. Meta has reiterated that it is not training models on private communications, maintaining privacy boundaries in line with its public stance.
“This training helps improve the performance, inclusiveness, and safety of the AI experiences we’re building. It also helps us better understand the languages, geography and cultural references that are relevant in Europe.”
AI Advancement for European Communities
Meta argues that training its AI models on public European content is essential to ensure that its generative AI tools reflect local languages, dialects, humor, and regional knowledge. With multi-modal AI functionality, spanning text, voice, video, and imagery now more advanced, Meta says such region-specific training is increasingly necessary.
This also marks a significant expansion for Meta AI, the company’s suite of generative tools now embedded across its platforms. The chatbot-like assistant, rolled out last month across Facebook, Instagram, WhatsApp, and Messenger in the EU, is designed to support everything from search queries and stickers to real-time chat assistance.
Notifications & Opt-Out Available to Users
Starting this week, Meta users across the EU (as well as in the EEA and UK) will begin receiving in-app and email notifications explaining:
- What kind of public data Meta will use
- How the data will improve Meta AI
- A link to an objection form, allowing users to opt out at any time
“We’ve developed a simple and accessible form that anyone in the EU, EEA or UK can use to object to our use of their information for AI training. We will honor all objections, including those we’ve already received.” Meta stated.
This opt-out mechanism is part of Meta’s pledge to comply with European data laws, offering a user-first approach amid ongoing scrutiny over AI ethics and data rights.
A Response to Growing Global Competition
Meta’s AI training approach mirrors similar strategies by other tech giants. The company pointed out that Google and OpenAI have already trained their AI models using data from European users, and stressed that Meta’s rollout is more transparent and compliant with GDPR.
In parallel, Apple recently published a report outlining how it uses differential privacy and synthetic data generation to train its AI features without compromising user privacy, signaling that AI development in Europe is becoming a battleground for ethical innovation.
“We believe we have a responsibility to build AI that’s not just available to Europeans, but is actually built for them. That means understanding the distinct ways different countries use humor, dialects, and sarcasm on our platforms.” Meta said in a blog post.
Regulatory Approval and What Comes Next
Meta had previously paused its AI training efforts in 2023 amid legal uncertainty. Now, with the EDPB’s December opinion affirming compliance and ongoing dialogue with the IDPC, Meta says it is confident its revised approach meets all relevant regulations.
The company also confirmed that it will continue engaging with regulators and provide updates as the AI tools evolve.
“Our strategy and plans will not change as a result of recent developments,” Meta said, reaffirming its commitment to long-term AI innovation in Europe.
Why It Matters
Meta’s move signals a broader trend among tech giants to tailor AI for local relevance while navigating stringent data laws. For Europeans, the decision marks both an opportunity for AI inclusion and a reminder of the importance of data transparency and user consent.
With the launch of Meta AI across Europe and the renewed AI training effort, the company appears poised to scale its generative tools rapidly, while balancing privacy and innovation.
