The CEO of OpenAI warns that the company could withdraw its services from the European market in response to the AI Act developed by the EU: "If we can’t comply, we will leave Europe".
OpenAI CEO Sam Altman explained that the startup could leave the European market if it does not find a "square" with EU regulations, i.e. the European AI Act.
Altman’s words leave no room for doubt and sound like a real bye-bye to the EU.
The mutual distrust between OpenAI and the European authorities has been known for some time and began with the provisional ban of ChatGPT in Italy, leading to direct monitoring of ChatGPT in Spain and the creation of a task European supervisory force which should lead to the approval of the AI European Act, which is currently still being finalized by the Brussels legislators.
According to The Verge, he reports that the new European regulation would worry Sam Altman a lot, who explained to some journalists, after a talk in London, that he had "many concerns" about the EU’s AI Act during a talk in London.
The terms of the law have been expanded in recent months to include new obligations for creators of so-called "base models," large-scale AI systems that power services like ChatGPT and OpenAI’s DALL-E.
"Details really matter," Altman said, according to a Financial Times report. "We will try to adapt to European requests, but if we cannot do so, we will simply stop operating in Europe ".
In comments reported by Time Magazine, Altman said the concern was that systems like ChatGPT would be designated as "high risk" under EU law. This means that OpenAI would have to meet a number of security and transparency requirements.
"Either we will be able to solve these requirements or not," said Altman who concluded: "there are technical limits here to what is possible."
If this classification is approved, OpenAI should modify the security and transparency parameters of its Chatbot for all European users. Altman, in this regard, reported that "we may be able to make the changes that may be required of us, or we may be unable to do so. There are technical limits to what can be done.
Another problem, according to Altman, would be the possible obligation for OpenAI to unveil the design of the ChatGPT system: the latter, in fact, was shared with the public until March, but when the Chatbot became economically valuable OpenAI has stopped publishing this data, fearing competition from other companies.
Furthermore, there is always the question of the possible (and as yet unconfirmed) use of copyrighted information for ChatGPT training, which, if openly disclosed to the public, would risk dragging OpenAI into huge legal trouble.
Original article published on Money.it Italy 2023-05-29 10:00:00. Original title: Sam Altman attacca le regole UE sulla IA e minaccia: ChatGPT via dall’Europa
