Meta AI expands in Europe despite regulatory hurdles
Meta announced on Thursday that the Meta AI chatbot will be released in 41 European countries starting this week. In mid-2024, after intervention from the Irish regulator, Meta halted plans to deploy this tool.
The company announced this week that the Meta AI assistant will be available in more than 60 markets, including 41 European countries. This is the largest expansion of Meta AI availability worldwide to date.
As communicated, Meta AI's first feature will be an intelligent chatbot, which will be available "for now" in six European languages: English, French, Spanish, Portuguese, Italian, and German. The chatbot will be free on Facebook, Instagram, WhatsApp, and Messenger.
Meta explained that users can engage with it through direct messaging to ask questions, explore interesting topics, or seek help solving problems. The company added that the assistant can also be used in group chats—initially on WhatsApp. Eventually, the feature will be introduced on Messenger and for private messages on Instagram.
It was announced that Meta AI will also be integrated with platform search bars, allowing it to provide "best-matching" results from various sources, such as reels, photos, and friends' and followers' posts.
Meta stated that the goal is to make its artificial intelligence tool available "on par" with the U.S., and it intends to "gradually" expand its offering in European countries.
Putting Meta AI technology into the hands of Europeans took significantly longer than the company planned. The reason for the delay was the European Union's complex regulatory system, stated the big tech company.
In the first half of 2024, Meta notified selected Facebook and Instagram users about the new AI policy and the possibility of using, for example, users' posts for training. At that time, the Irish Data Protection Commission (DPC) reacted (Meta's European activities are registered in Ireland), highlighting concerns about training AI on European users' data obtained from posts, photos, and captions published on Facebook and Instagram.
Meta offered the possibility to express objections to the new policy, but a special form (opt-out) had to be completed for this purpose. Some experts, including the non-governmental organization Noyb that deals with data protection, argued that such a practice contradicts GDPR. Noyb called the opt-out form offered by Meta "a farce." They also criticized that Meta retained the right to disregard objections.
The organization emphasized that big tech practices included many "dormant" accounts on Facebook, with which users hardly interacted anymore but which still contained vast amounts of personal data. Noyb filed 11 complaints related to Meta AI in June 2024 to data protection authorities in various EU countries, including Poland’s Personal Data Protection Office (UODO).
Implementation of the new tool in the EU faces challenges
After intervention by the Irish Data Protection Commission, Meta halted the training of AI on information from Facebook and Instagram users in the European Union. The Irish Commission, representing several European authorities in the Meta AI matter, expressed satisfaction with this decision in a statement. It was simultaneously announced that talks with the big tech companies would continue.
"We’re disappointed by the request from the Irish Data Protection Commission (DPC), our lead regulator, on behalf of the European DPAs, to delay training our large language models (LLMs) using public content shared by adults on Facebook and Instagram — particularly since we incorporated regulatory feedback and the European DPAs have been informed since March," stated Meta in a statement released in early June 2024. The company then noted that Meta AI "at this point" will not be able to be deployed in the European Union.
"We are honouring all European objections. If an objection form is submitted before Llama training begins, then that person’s data won’t be used to train those models, either in the current training round or in the future," assured Meta in June 2024.
Meta Platforms is an American corporation that owns social media and communication platforms: Facebook, Instagram, Messenger, and WhatsApp. Meta AI is the research division of Meta Platforms that develops artificial intelligence and augmented reality (AR) technologies. In 2023, Meta AI launched a family of large language models under the name Llama, which have been updated several times since then.
