Microsoft splits with Meta on EU’s voluntary AI compliance framework by agreeing to it

Microsoft will likely sign the European Union’s voluntary AI code of practice to help companies comply with the bloc’s landmark artificial intelligence rules, while Meta Platforms has rejected the guidelines. The code, developed by 13 independent experts, aims to provide legal certainty for AI companies by requiring them to publish summaries of training content and establish copyright compliance policies.

What you should know: The code of practice is part of the EU’s AI Act, which came into force in June 2024 and will apply to major tech companies including Google, Meta, OpenAI, Anthropic, and Mistral.

• Companies that sign the voluntary code must publish summaries of the content used to train their general-purpose AI models and implement policies to comply with EU copyright law.
• The guidelines are designed to help thousands of companies navigate compliance with the bloc’s comprehensive AI regulations.

The big picture: Major AI companies are taking divergent approaches to EU compliance, with some embracing the voluntary framework while others resist what they see as regulatory overreach.

• OpenAI and Mistral have already signed the code of practice, demonstrating early adoption among some AI developers.
• The split response highlights ongoing tensions between tech companies and European regulators over AI governance.

What they’re saying: Microsoft President Brad Smith expressed cautious optimism about joining the voluntary framework.

• “I think it’s likely we will sign. We need to read the documents,” Smith told Reuters on Friday.
• “Our goal is to find a way to be supportive and at the same time one of the things we really welcome is the direct engagement by the AI Office with industry,” he said, referring to the EU’s regulatory body for AI.

Why Meta is pushing back: The social media giant argues the code introduces unnecessary legal uncertainties and exceeds the scope of the AI Act.

• “Meta won’t be signing it. This code introduces a number of legal uncertainties for model developers, as well as measures which go far beyond the scope of the AI Act,” Meta’s chief global affairs officer Joel Kaplan said in a LinkedIn blog post on Friday.
• Meta shares concerns with 45 European companies that the guidelines will “throttle the development and deployment of frontier AI models in Europe, and stunt European companies looking to build businesses on top of them,” according to Kaplan.