Experts calling for stricter AI rules

For the second time in a short time, experts are calling for stricter rules for artificial intelligence (AI). What are they afraid of and what is going on at the moment?
What’s going on?

A group of 50 AI experts and institutions wants European politicians to come up with additional rules for ‘general purpose AI’ or GPAI. That vague term includes artificial intelligence that was created without a specific purpose and can be used for multiple things.

Think, for example, of the immensely popular chatbot ChatGPT: a generative model that can produce texts itself, but can also check programming code for errors. Or Dalle – E and Midjourney, who can create works of art or photorealistic pictures based on words. Think of the photos where it looks like the pope is suddenly walking around in a white puffer jacket.

The fact that the experts are ringing the bell right now is not just that. The European Union has been working on rules for AI for two years and is now thinking and negotiating how strict the rules should be and what kind of AI applications are covered by the regulations.

But these critics fear that Brussels is using too narrow a definition for the rules that are now being put on paper.

Artificial intelligence models designed without a specific purpose should become part of the rules as far as they are concerned. They warn that what we see now is just the tip of the iceberg.

The companies that underpin the AI models must be transparent and held accountable for the data they use and the design choices they make, the experts said. They argue for rules around the development of the models, how data was collected, who participated in it and how the models were trained.

Regulations in which you determine in advance whether something is risky or not are not enough. Because applications that do not pose a risk in the initial phase can still do so later.