The EU has opened the process of drafting the code of practice to companies, academics, and other stakeholders, receiving nearly 1,000 applications
The world’s leading tech companies are ramping up efforts to convince the European Union (EU) to adopt a lighter approach in regulating artificial intelligence (AI). These companies are concerned about the potential for billions of dollars in fines if the rules are too stringent.
In May, the EU finalized the AI Act, marking the first comprehensive global framework to regulate AI technology. However, the details of how the law will be enforced, particularly concerning “general purpose” AI systems like OpenAI’s ChatGPT, remain uncertain. The final guidelines, including the code of practice, are still under development, which leaves the tech industry in limbo about the potential legal and financial implications they may face.
The EU has opened the process of drafting the code of practice to companies, academics, and other stakeholders, receiving nearly 1,000 applications—a high number, according to sources. Although the code will not be legally binding, it will act as a checklist for companies to demonstrate their compliance with the AI Act. Companies claiming to follow the law while ignoring the code could face legal action.
Boniface de Champris, a senior policy manager at the trade organization CCIA Europe, which represents companies like Amazon, Google, and Meta, highlighted the importance of getting the code right. “If we get it right, we will be able to continue innovating,” he said. “If it’s too narrow or too specific, that will become very difficult.”
One major concern for the tech industry is the issue of data scraping—using vast amounts of content such as books and photos to train AI models, often without permission. Stability AI and OpenAI have faced scrutiny for using copyrighted materials in this way, raising questions of whether this constitutes a breach of copyright laws.
Under the AI Act, companies will need to provide “detailed summaries” of the data used to train their AI models. This means content creators could seek compensation if they discover their work was used without permission, although this is currently being tested in courts. Some business leaders argue that these summaries should include minimal details to protect trade secrets, while others believe copyright holders have the right to know how their content is being used.
OpenAI, which has previously refused to disclose information about the data used in its models, has applied to join the working groups tasked with drafting the code. Google has also submitted an application, and Amazon expressed hopes to “contribute our expertise and ensure the code of practice succeeds.”
Maximilian Gahntz, AI policy lead at the Mozilla Foundation, voiced concerns over a lack of transparency. “The AI Act presents the best chance to shine a light on this crucial aspect and illuminate at least part of the black box,” he said.
Some business leaders have criticized the EU for focusing too much on regulating tech companies instead of encouraging innovation. The challenge for those drafting the code of practice is to find a balance between regulation and allowing room for technological growth.
Last week, Mario Draghi, former chief of the European Central Bank, urged the EU to improve coordination on industrial policy and speed up decision-making to stay competitive with China and the U.S. This call for faster action reflects growing concerns within Europe about keeping pace with global tech giants.
Thierry Breton, a strong advocate for strict tech regulation and critic of non-compliant companies, recently stepped down as European Commissioner for the Internal Market following a clash with EU Commission President Ursula von der Leyen.
Meanwhile, smaller European tech companies are pushing for exemptions in the AI Act to support startups. Maxime Ricard, policy manager at Allied for Startups, argued that these regulations should be “manageable” and tailored to the needs of smaller firms.
The finalized code is expected to be published in early 2025, with companies having until August 2025 to ensure their compliance. Non-profits like Access Now, the Future of Life Institute, and Mozilla have also applied to participate in drafting the code, alongside industry giants.
“As we enter the stage where many of the AI Act’s obligations are spelled out in more detail, we have to be careful not to allow the big AI players to water down important transparency mandates,” Gahntz warned.
In the coming months, the EU will need to strike a delicate balance between encouraging innovation and enforcing regulations that protect the rights of content creators and ensure AI development remains transparent.
