Skip to content

Meta initiates artificial intelligence education in Germany

Meta initiates artificial intelligence education in Germany

Large-scale Artificial Intelligence training on German user data initiated by Meta, reveled in...
Large-scale Artificial Intelligence training on German user data initiated by Meta, reveled in image archive.

Meta initiates Artificial Intelligence learning in Germany - Meta initiates artificial intelligence education in Germany

Meta Commences Large-Scale AI Training with German User Data Following Court Green Light

Meta, the parent company of Facebook and Instagram, has been given the go-ahead by a German court to use user data from both platforms to train its AI models on a broad scale. The planned AI training involves reviewing all posts ever made on Facebook and Instagram, but users can actively opt out before the deadline on Tuesday.

WhatsApp, another Meta-owned service, is not part of the training as chats are encrypted. However, Meta AI chats are considered public and could be utilized for training purposes.

Legal Hurdles Overcome

The German court in Cologne handed down a verdict earlier this month allowing Meta to employ user posts from Facebook and Instagram for AI software training (Case No. 15 UKl 2/25). The Higher Regional Court determined that Meta's use of user data in this manner did not breach EU law.

The consumer protection agency of North Rhine-Westphalia initiated the lawsuit, alleging violations of European data protection law. After the court ruling, a Meta spokesperson maintained that their AI training approach did not infringe any data protection regulations and aimed to provide equal access to advanced AI technology for the entire European population.

Competitive Landscape

In a recent study by Bitkom, an association representing Germany's digital industry, Meta trailed behind competitors like OpenAI (ChatGPT), Microsoft (CoPilot), and Google (Gemini) in the adoption of AI applications by German users. Meta's AI language model Llama did not appear in the list of most popular AI applications.

Following its introduction in 2023, Llama garnered attention within expert circles as an open-source system. However, many experts now doubt Meta's ability to keep pace with the market leaders, which includes US giants, as well as emerging Chinese competitors like DeepSeek.

Background: Court Ruling and Opt-Out Period

The German court ruled in favor of Meta, allowing the company to use user data for AI training, concluding that Meta did not violate EU law and justifying its AI data processing under the legitimate interest clause of GDPR (Article 6(1)(f)). The court emphasized that Meta's approach ensured compliance with data protection standards by providing transparency and opt-out options to users. The ruling will hold consequences for data protection in Germany and the EU, setting precedents for AI training-related data processing under GDPR.

Meta's AI training expansion now includes user data from EC countries, following the court ruling in Cologne (Case No. 15 UKl 2/25). This move places the company's employment policy of AI development in alignment with the diverse teams of users across Europe.

Germany's digital industry, represented by Bitkom, reported that Meta trails behind competitors in AI adoption by local users. This comes as Meta's new AI language model Llama was introduced in 2023, yet questioned by experts regarding its ability to compete with market leaders in finance, technology, data-and-cloud-computing, and artificial-intelligence.

In light of the court ruling, Meta must ensure transparency and provide opt-out options to users, ensuring their employment policy on data protection adheres to the highest standards, aligning with the ever-evolving landscape of business and industry.

Read also:

    Latest