Meta announced a postponement of its AI assistant's launch and training in Europe following a request from the Irish Data Protection Commission (DPC). The regulator asked Meta to delay training its large language model (LLM) using public content shared by adults on Facebook and Instagram, as stated in a June 10 release. Acting as Meta’s primary privacy regulator in the EU, the Irish DPC made the request on behalf of European Data Protection Authorities (DPAs).

Due to insufficient local information, Meta indicated that European users would receive a subpar experience, leading to the current inability to launch Meta AI in Europe. Meta expressed disappointment in the decision, viewing it as a setback for AI competition and innovation within the EU, though it maintains confidence in its compliance with European regulations.

The Irish DPC supported the pause in training activities and expressed satisfaction with Meta’s decision, continuing their collaboration following earlier extensive engagement. The decision to halt Meta’s activities came after complaints from various entities, including the European privacy non-profit NOYB, the Norwegian Consumer Council, and specific DPAs, prompting action from the Irish DPC.

Meta disclosed that it had been informing European DPAs about its activities since March and had integrated feedback following consultations with the Irish DPC by June 10. Despite the setback, Meta highlighted that other tech companies like Google and OpenAI also utilize AI for user data training. Meta anticipates aligning its AI training practices with EU GDPR laws, leveraging the legal basis of "Legitimate Interest," similar to practices observed elsewhere in the AI industry.