The European Union on Thursday released a code of practice on general purpose artificial intelligence to help thousands of businesses in the 27-nation bloc using the technology comply with the bloc’s landmark AI rule book.
The EU code is voluntary and complements the EU’s AI Act, a comprehensive set of regulations that was approved last year and is taking effect in phases.
The code focuses on three areas: transparency requirements for providers of AI models that are looking to integrate them into their products; copyright protections; and safety and security of the most advanced AI systems.
The AI Act’s rules on general purpose artificial intelligence are set to take force on Aug. 2. The bloc’s AI Office, under its executive Commission, won’t start enforcing them for at least a year.
General purpose AI, exemplified by chatbots like OpenAI’s ChatGPT, can do many different tasks and underpin many of the AI systems that companies are using across the EU.
Under the AI Act, uses of artificial intelligence face different levels of scrutiny depending on the level of risk they pose, with some uses deemed unacceptable banned entirely. Violations could draw fines of up to 35 million euros ($41 million), or 7% of a company’s global revenue.
Some Big Tech companies such as Meta have resisted the regulations, saying they’re unworkable, and U.S. Vice President JD Vance, speaking at a Paris summit in February, criticized “excessive regulation” of AI, warning it could kill “a transformative industry just as it’s taking off.”
More recently, more than 40 European companies, including Airbus, Mercedes-Benz, Philips and French AI startup Mistral, urged the bloc in an open letter to postpone the regulations for two years. They say more time is needed to simplify “unclear, overlapping and increasingly complex EU regulations” that put the continent’s competitiveness in the global AI race at risk.
There was no sign that Brussels was prepared to stop the clock.
“Today’s publication of the final version of the Code of Practice for general-purpose AI marks an important step in making the most advanced AI models available in Europe not only innovative but also safe and transparent,” the commission’s executive vice president for tech sovereignty, security and democracy, Henna Virkkunen, said in a news release.
Learn More About Securing AI at SecurityWeek’s AI Risk Summit – August 19-20, 2025 at the Ritz-Carlton, Half Moon Bay
Related: The Wild West of Agentic AI – An Attack Surface CISOs Can’t Afford to Ignore
Related: What Can Businesses Do About Ethical Dilemmas Posed by AI?
