EU Lays Out AI Guidelines for Companies --- Commission and member states have to decide if they will endorse the code

 

“The European Union said it published the final version of a voluntary code of practice for general-purpose artificial intelligence that is meant to make it easier for companies to comply with AI legislation in the bloc.

 

The European Commission, the EU's executive arm, said the code was developed by 13 independent experts, based on input from more than 1,000 small and medium-size enterprises, academics, civil society groups and providers of the large language models that underpin AI.

 

Now, the commission and the EU's 27 member states will have to sift through the code and decide whether to endorse it. Model providers will then have the option to sign the code, which EU officials said would help companies to comply with the bloc's wide-ranging legislation on the technology.

 

EU lawmakers approved the AI Act last year. The law bans certain uses of AI, rolls out new transparency guidelines and requires risk assessments for AI systems that are deemed high-risk.

 

Rules on general-purpose AI will be effective for companies as of Aug. 2. The commission said its AI office, a body that oversees implementation of the law, would enforce rules on new AI models after a year, and two years later for existing models. Companies that breach the law risk fines of up to 7% of their annual global revenue.

 

The code includes guidance on safety and security to help companies mitigate risks when developing advanced AI models, a section on copyright material as well as a chapter on transparency.

 

Signatories to the code of practice will benefit from a lower administrative burden and extra legal certainty to prove they comply with the AI Act compared with companies that opt not to sign the code. Henna Virkkunen, EU executive vice president for tech sovereignty, security and democracy, said signing the code would secure a clear route to compliance with the law.

 

A spokesperson for OpenAI said the ChatGPT maker was reviewing the code before deciding whether to sign it. News Corp, owner of Dow Jones Newswires and The Wall Street Journal, has a content-licensing partnership with OpenAI.

 

A spokesperson for Alphabet's Google said the company would review the code. "Europeans should have access to first-rate, secure AI models when they become available, and an environment that promotes innovation and investment."

 

The EU's push to regulate AI has drawn criticism from business leaders and politicians alike. Chief executives of European companies, including Mistral AI, ASML Holding and Airbus, recently asked the commission to delay enforcement of the law, saying overlapping and complex regulations were preventing the EU from becoming a leader in AI.

 

Earlier this year, U.S. Vice President JD Vance told world leaders gathered in Paris for an AI summit that "excessive regulation of the AI sector could kill a transformative industry just as it's taking off."

 

Commission officials acknowledged that regulatory fragmentation across the EU's 27 countries and inadequate financing were holding back growth for startups. The EU in May laid out plans to slash regulation and set up a special fund to attract tech startups as it plays catch-up with the U.S. on financing and innovation.” [1]

 

1. EU Lays Out AI Guidelines for Companies --- Commission and member states have to decide if they will endorse the code. Orru, Mauro.  Wall Street Journal, Eastern edition; New York, N.Y.. 11 July 2025: B4.