Meta initiates AI learning utilizing user data in their systems.
Germany's Higher Regional Court has given Meta the green light to train its AI systems using user data from Facebook and Instagram. The US tech giant can now proceed with reading out all posts made by users on these platforms to improve its AI models.
Initially, users were able to object to the use of their data, but the deadline for objections has lapsed. WhatsApp, another Meta-owned platform, is excluded from the AI training as chats are encrypted. However, public chats with Meta's AI assistant, "Meta AI," can be used for this purpose.
Critics, including Verbraucherzentrale NRW, had sought to prevent Meta from using user data for AI training, arguing it violated European data protection laws. The court dismissed the argument, with the judge stating that the AI training process did not breach any data protection regulations.
Meta emphasized their commitment to developing AI based on data from European users, ensuring equal access to its benefits across the region. However, concerns about data privacy persist. The company has faced criticism for not providing clear information about how user data is used and for violating users' rights to object to the data processing.
Legal proceedings against Meta are ongoing, with the Hamburg Data Protection Commissioner seeking an urgent three-month halt to AI training in Germany. European-wide legal challenges may also emerge, with organizations like noyb contemplating pan-European actions.
The General Data Protection Regulation (GDPR) governs how Meta uses user data for AI training in the EU. Key GDPR principles, such as consent, transparency, and fairness, are at the heart of the concerns raised by critics. The legal landscape remains dynamic, with potential future litigation that could impact Meta’s AI development plans.
Sources: ntv.de, gho/dpa
In light of the ruling by Germany's Higher Regional Court, Meta's AI models will now be trained using social policy data from Facebook and Instagram, potentially involving industry, finance, technology, and artificial-intelligence sectors. Despite the court's dismissal of privacy concerns, Verbraucherzentrale NRW and other critics continue to contest Meta's use of user data for AI training, citing violations of European data protection laws such as the GDPR's principles of consent, transparency, and fairness.
