Intellience

Microsoft signs up for European Union’s AI code of conduct

Redmond signs up and Meta opts out as legislators say there will be no changes to AI Act timeline
Pro
Image: Stockfresh

24 July 2025

Microsoft executive Brad Smith has indicated that the company is likely to participate in the European Union’s code of conduct for artificial intelligence (AI). The voluntary code is intended to provide legal clarity for companies seeking to comply with the EU’s groundbreaking AI regulations. Signatories must publish summaries detailing the data used to train their general-purpose AI models. In addition, they must establish policies that comply with EU copyright law.

The code is an integral part of the AI Act, which came into force in June 2024. The law applies to a wide range of companies, including giants such as Alphabet (Google’s parent company), Meta (Facebook’s parent company), OpenAI, Anthropic, Mistral, and thousands of others. Smith emphasised Microsoft’s willingness to cooperate with the EU’s AI regulator.

The EU code is designed to help companies implement processes and systems to comply with AI regulations. The main requirements are to provide and regularly update documentation on AI tools and services, prohibit training on illegal content, and require companies to comply with rights holders’ requests regarding the use of their works in datasets.

 

advertisement



 

Meta, on the other hand, has decided not to sign the code. Joel Kaplan, Meta’s head of global affairs, expressed his concerns. According to him, the code creates legal uncertainty for developers and goes beyond what the AI legislation requires.

He criticised the European Union’s approach, saying it would hinder the development of AI and innovation in Europe.

The AI Act itself is a risk-based regulatory framework for AI applications. It bans certain uses with an ‘unacceptable risk’, such as manipulating cognitive behaviour and social scoring. The law also defines ‘high-risk’ applications, including biometrics and facial recognition, in areas such as education and employment. Developers must register AI systems and comply with obligations regarding risk and quality management.

Despite opposition from tech giants such as Alphabet, Meta, Microsoft, and Mistral AI, the European Union is sticking to its timeline for implementing the AI Act.

Business AM

Read More:


Back to Top ↑

TechCentral.ie