Any individual who had intended to develop services utilizing Meta’s AI models may be prevented from selling those services in Europe as a result of this move.
According to a report by Axios, Meta has made the decision to not provide its forthcoming multimodal artificial intelligence model and any future versions to consumers in the European Union. The company cites a lack of clarification from European regulators as the reason for this decision. The aforementioned models are intended to process not just text but also images and audio, and they are also responsible for powering artificial intelligence capabilities in Meta platforms as well as the Ray-Ban smart glasses that the business manufactures.
“We will release a multimodal Llama model over the coming months, but not in the EU due to the unpredictable nature of the European regulatory environment,” Meta stated in a statement made available to Axios.
Meta’s decision comes after Apple made a decision that was quite similar to this one. Apple recently announced that it will not release its Apple Intelligence features in Europe owing to worries over regulatory compliance. Apple’s decision was criticized by Margrethe Vesteger, the competition commissioner for the European Union. She stated that the company’s choice was a “stunning, open declaration that they know 100 percent that this is another way of disabling competition where they have a stronghold already.” Vesteger has been contacted by Engadget in an effort to obtain a response on Meta’s decision.
Should the European Union decide to withhold Meta’s multimodal artificial intelligence models from its jurisdiction, it might have far-reaching ramifications. This would mean that any businesses that use these models to construct their goods and services would be unable to sell them in Europe.
Meta has informed Axios that it is still planning to distribute Llama 3, the forthcoming text-only model designed by the business, in the European Union. The issues that arise from training artificial intelligence models using data from European clients while adhering to the General Data Protection Regulation (GDPR), which is the regulation that governs data protection in the European Union, are the key source of concern for the organization. For the purpose of training future artificial intelligence models, Meta made an announcement in May that it intended to use publicly available posts from users of Facebook and Instagram. However, the company was obliged to stop doing so in the European Union (EU) after receiving opposition from data privacy officials in the region. Meta provided an explanation for its actions at the time, stating that it was essential for the company to be able to train its models on the data of European customers in order to accurately reflect the local culture and terminology.
“If we don’t train our models on the public content that Europeans share on our services and others, such as public posts or comments, then models and the AI features they power won’t accurately understand important regional languages, cultures or trending topics on social media,” according to a blog post published by the organization. “We believe that Europeans will be ill-served by AI models that are not informed by Europe’s rich cultural, social and historical contributions.”
While Meta has some reservations about releasing their multimodal models in the European Union (EU), the company still intends to offer them in the United Kingdom (UK), which has data privacy rules that are comparable to those of the EU. It was stated by the corporation that European authorities are taking significantly more time to interpret the regulations that are already in place in comparison to their counterparts in other places.