News Hub

Meta announces plan to train AI on European social media posts

Written by Thu 13 Jun 2024

Facebook parent company, Meta, has announced they are seeking to use social media posts in the European Union to train its own generative artificial intelligence (GenAI) model.

The Meta-owned Llama large language model will be fed content from people in the European Union who have opted to share posts publicly on Instagram and Facebook, according to a blog post.

“If we do not 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 will not accurately understand important regional languages, cultures, or trending topics on social media,” said the blog post.

However, due to comprehensive data privacy laws in the European Union, it may be a challenge for the social media giant to carry out these plans without opposition. Austrian non-profit organisation NOYB – European Center for Digital Rights have already send complains about this plan to almost a dozen privacy watchdogs across Europe to demand Meta are stopped from training its model on user data.

Both Google and OpenAI have already used content from users in the European Union to train AI models, with Meta saying they intend to take a more transparent approach than other social media firms.

Meta Consults with Privacy Regulator 

Meta have made it clear that private messages and content from under-18 year olds will not be used during this project to train its generative AI model. To ensure local laws and regulations are taken into consideration, Meta are consulting with its lead privacy regular in the EU, the Irish Data Protection Commission.

The updated version of Meta’s privacy policy will come into effect on 26 June, with training expected to happen soon after this data.

In a bid to promote responsible AI development, in December Meta and IBM launched an AI Alliance to advance innovation and open science in AI.

The AI Alliance comprises 50 leading organisations across industry, startup, academia, research, and government. Members of the Alliance include AMD, CERN, Dell Technologies, Intel, Oracle, NASA, Sony and more. The Alliance is dedicated to fostering an open community to assist developers and researchers in accelerating responsible AI innovation.

In October, Google DeepMind, Anthropic, OpenAI, Microsoft, Amazon, and Meta published AI safety policy updates following pressure from the UK Government. The technology companies hope to boost transparency and encourage the sharing of best practices within the AI community.

Join Tech Show Paris

27-28 November 2024, Porte de Versailles, Paris

Be a part of the latest tech conversations and discover pioneering innovations in Paris.

Don’t miss one of the most exciting technology events of the year for France.

Written by Thu 13 Jun 2024

Send us a correction Send us a news tip