Meta is bringing Meta AI to the European Union, but the launch is not the same product users have seen in the U.S. The assistant is arriving as a narrower chat tool, shaped by the company’s continuing clash with European privacy regulators over how user data can be used to train AI models.
The rollout covers all 27 EU countries, plus an additional 14 European countries and 21 overseas territories. Meta is also adding the assistant to WhatsApp in the U.K., where it had previously been available through Facebook, Instagram, and Ray-Ban Meta glasses since its launch there in October.
A broad rollout with a limited feature set
Meta AI has been available in the U.S. since 2023. In that market, the assistant can chat, answer questions, generate images, and create stylistic selfies, along with other creative features.
The European launch is more restrained. For now, Meta AI in the EU is limited to what the company calls an "intelligent chat function" in six European languages: English, French, Spanish, Portuguese, German, and Italian.
The assistant will appear inside WhatsApp, Instagram, Messenger, and Facebook. Users can summon it by tapping a small blue circle icon and asking the kind of question they might otherwise type into a search engine, such as how to complete a task or where to find information on a topic.
Meta is also bringing the assistant into group chats. That rollout will be staggered, beginning with WhatsApp in both the EU and the U.K. Meta says Messenger and Instagram Direct Messaging will follow "soon."
In group chats, users will be able to call the assistant by typing "@MetaAI" before asking a question. The examples given include asking where to go for dinner or identifying the top tourist attraction in a city.
Why Europe is different
The smaller launch reflects the regulatory pressure Meta has faced in Europe. The company has trained its AI on user-generated content in the U.S. for years, but the EU’s privacy framework creates a different standard for processing people’s information.
The source of the tension is the General Data Protection Regulation (GDPR). Under that regime, Meta needs a valid legal basis if it wants to use personal information to train AI models.
Last May, Meta began notifying regional users about an upcoming privacy policy change. The change said the company would begin using content from comments, interactions, status up dates, photos, and captions for AI training.
Meta argued that this use of data was necessary so its AI model could reflect “the diverse languages, geography and cultural references of the people in Europe.” But in June, the company had to pause those plans after scrutiny from the Irish Data Protection Commission (DPC), Meta’s lead data protection regulator in the EU.
The concern centered on the way Meta was asking users to object. Meta relied on an opt-out process, meaning users had to take action to stop their data from being used for AI training. The company said it was relying on a GDPR legal basis known as “legitimate interests,” but the DPC disagreed and Meta had to rethink its approach.
Meta says EU user data was not used
For this launch, Meta says the assistant being released in the EU was not trained on local users’ data. That claim is central to how the company is presenting the rollout.
Because Meta says the technology was not trained on EU users’ information, it also says it will not notify users or seek consent for this version of the product. In Meta’s framing, the current assistant is a chat product arriving without the data-training issue that stalled its earlier plans.
"The model powering these Meta AI features wasn’t trained on first-party data from users in the EU," Anna Dack, Meta’s innovation communications manager, EMEA, told TechCrunch.
That distinction helps explain why the European Meta AI feature set is narrower than the U.S. version. Image generation, stylistic selfies, and other creative features are not part of the EU release yet.
Meta’s own language also suggests this is not the end state. The company calls the launch a "first step" in bringing more AI to Europe and says it wants to eventually "find parity with the U.S." over time.
The U.K. path shows the regulatory split
The U.K. sits outside the EU since Brexit, but its data protection regime is still based on the GDPR. Meta faced similar concerns there over its AI training plans.
Last summer, the U.K.’s Information Commissioner’s Office (ICO) asked Meta to pause its AI training plans because of concerns about how the company was using user data. Meta later adjusted its opt-out process, making it mildly less onerous, and then launched Meta AI in the U.K.
The ICO did not explicitly object to that U.K. launch, though it said it would "monitor the situation." When asked whether Meta’s AI efforts had been trained on U.K. users’ data, a spokesperson pointed TechCrunch to Meta’s September announcement saying it would begin training its models on user content "in the coming months."
That answer suggested the company was not yet ready to put that part of its AI strategy on the public stage.
What to watch next
The European Meta AI launch is important because it shows Meta moving ahead while trying to stay inside the boundaries set by regulators. The product is arriving, but without the full U.S. feature set and without Meta claiming it used first-party EU user data to train the model behind these features.
The DPC is still watching. TechCrunch asked the regulator for its response to the EU announcement, and the DPC said it has been examining Meta AI over recent months with other Supervisory Authorities across the EU/EEA.
“The DPC, as Lead Supervisory Authority for Meta, has been examining Meta AI over recent months with our colleague Supervisory Authorities across the EU/EEA and we will keep it under review as it rolls-out to users over the coming weeks," a spokesperson told us.
For users, the immediate change is practical: Meta AI will become available inside familiar apps as a chatbot that can answer questions in one-to-one and group conversations. For Meta, the larger issue remains unresolved: how far it can expand AI features in Europe while regulators continue to examine the role of user data in AI training.