Meta, formerly known as Facebook, has hit a roadblock in its plans to launch an AI assistant in Europe. The Irish Data Protection Commission (DPC) has requested Meta to pause the training of its large language model (LLM) involving public content shared by adults on Facebook and Instagram. This delay comes after complaints from various organizations and DPAs, including NOYB and the Norwegian Consumer Council.
Meta expressed disappointment in the request, calling it a setback for AI competition and innovation in the EU. The company believes it complies with European regulations but has agreed to postpone the launch of its AI assistant in Europe for the time being. The lack of local information for training means that users would not receive the optimal experience.
The Irish DPC, acting as Meta’s lead privacy regulator in the EU, supports the decision to pause the AI training. The commission will continue to work with Meta after previous intense engagement on the matter. Despite the setback, Meta remains hopeful that it can eventually launch its AI assistant in Europe while complying with the EU’s General Data Protection Regulation (GDPR).
The situation has been developing over the past few months, with Meta informing European DPAs of its activities since March. However, the issue escalated after Meta notified users of its AI training activities through two billion in-app notifications and emails in May. This led to complaints from various organizations and ultimately the request from the Irish DPC to halt the training.
Meta has pointed out that other tech giants like Google and OpenAI already use AI for user data training. The company aims to comply with GDPR laws by citing “Legitimate Interest” as the legal basis for its AI training, a provision that others in the industry have also relied on.
This delay in launching the AI assistant in Europe reflects the challenges that tech companies face when navigating regulatory frameworks in different regions. While Meta remains committed to innovation and competition in the AI sector, it must also ensure compliance with privacy regulations to gain the trust of users and regulators alike.