A proposed EU law that would create the West’s first regulatory scheme for AI is the target of criticism from the business and technology sectors, which published an open letter today arguing for fewer restrictions and a less “bureaucratic” approach.
The letter was signed by prominent executives from a who’s-who of major European companies like Siemens, Dassault, Heineken, Renault, Deutsche Telekom, and Airbus. In total, 163 people signed the letter.
The signatories urged the EU to adopt a more hands-off approach to AI regulation, worrying that the draft AI Act would make the continent less competitive in the fast-growing field.
“Wanting to anchor the regulation of generative AI in law and proceeding with a rigid compliance logic is as bureaucratic of an approach as it is ineffective in fulfilling its purpose,” the letter said. “In a context where we know very little about the real risks, the business model, or the applications of generative AI, European law should confine itself to stating broad principles in a risk-based approach.”
The letter stressed the importance of generative AI, likening it to the invention of the microchip or the internet, and stated that the need to comply with the law could result in major AI innovators relocating out of Europe.
“Under the [draft law] recently adopted by the European Parliament, foundation models, regardless of their use cases, would be heavily regulated, and companies developing and implementing such systems would face disproportionate costs and disproportionate liability risks,” the letter said.
The AI Act, which passed the European Parliament earlier this month, will become law if and when it is ratified by each EU country. Provisions include a blanket ban on the use of AI in biometric identification, a requirement for labeling of AI-generated content, and safeguards against AI and illegal content.
The act was amended in April to include more regulation for generative AI, which prompted some debate over last-minute changes. Ultimately, lawmakers formed a consensus that large language models such as ChatGPT, Midjourney have to be regulated to preserve core EU rights and values, like freedom of expression. A provision that would require all such generative AI creators to disclose copyrighted material was also included.