Yuan Announce On Monday, it will train its AI model on public content such as Facebook and Instagram posts and comments, after which Suspend its plans Respond to regulatory pressures caused by data privacy issues. It said that training on EU users will begin this week. User interaction with meta-AI will also be used to train their models.
The announcement is Meta AI launches in the EU Last month, after debuting in the U.S. and other global markets.
And Yuan has Training its AI User-generated content generated in the United States has faced resistance in the EU due to the group’s strict privacy laws, especially the General Data Protection Regulation (GDPR), which requires clear legal basis for processing personal data to train AI models.
Meta says As early as June 2024 It will suspend plans to start training its AI systems using EU and UK user data after the pushback of the Irish Data Protection Commission (DPC). DPC regulates the EU’s dollar and acts on behalf of several data protection authorities across the group. exist September 2024Meta said it is restarting efforts to use public posts from the UK user base to train its AI systems.
Fast forward to today, and Meta announced it would do so in public posts of its EU user base.
“Last year, we delayed training large language models with public content, while regulators articulate legal requirements,” Mehta said in a blog post. “We welcomed the comment provided by EDPB in December that confirmed that our original approach was in line with our legal obligations. Since then, we have interacted constructively with IDPC and look forward to continuing to bring the full benefit of generating AI to people in Europe.”
Starting this week, EU users will begin receiving in-app and email notifications to explain that META will begin training its models using public data and interactions with Meta AI. These notifications will include a link to the form that will allow the user to opt out of the data used. Mehta said it will respect all objection forms it has received and the newly submitted forms.
Meta notes that it does not use private messages from EU users under the age of 18, nor does it use private messages to train its models.
“We believe it’s our responsibility to build AI not only for Europeans, but actually for them,” Mehta said. “That’s why it’s so important to train our generative AI models in a variety of data so that they can understand the incredible, diverse nuances and complexities that make up the European community. This means from dialect and colloquial knowledge to hyper-local knowledge, and the ways in which a variety of different countries use humor and irony on our products.”
Meta said this is an example set by companies such as Google and OpenAI, which have used data from European users to train their AI models.
Meanwhile, DPC does not fully review the ways large language model creators are training their AI services. Last week, regulators Announce It is investigating Xai’s training on Grok.
