She can write and speak like a human being. She creates images, formulates poems like an artist. She writes scientific essays and class tests. It assists doctors in detecting skin or breast cancer. It is part of war drones and other intelligent weapon systems. There seems to be almost nothing that artificial intelligence (AI for short) can't do. And it is becoming more and more powerful, which can be observed in the AI system ChatGPT from the American company Open AI. Many are frightened by the rapid progress. Twitter and Tesla CEO Elon Musk considers artificial intelligence potentially "more dangerous than the atomic bomb" and calls for a pause in development with other managers and scientists.

Manfred Lindinger

Editor in the "Nature and Science" department.

  • Follow I follow

What is still real, what is fake? What is helpful, what is harmful? What does the development of this technology mean for work, prosperity, democracy and for the question of war or peace? Do the Germans have too many concerns and are gambling away their future, into which the Americans and China have long since started uncatchable? Maybritt Illner asked these questions to her five guests on Thursday evening in her show "Artificial Intelligence – Machine vs. Human?": Ranga Yogeshwar, physicist and science journalist, Miriam Meckel, communication scientist and journalist, Anke Domscheit-Berg, net activist and politician of the party "Die Linke", Saskia Esken, party chairwoman of the SPD and trained computer scientist, and Achim Berg, president of the digital association Bitkom.

What was the first thing he asked ChatGPT, so the host asked Ranga Yogeshwar on her right right at the beginning of the show. It was like a rush for the physicist when ChatGPT emerged at the end of last year. "You tried everything. I asked the AI, 'Am I married?' And then it came out: 'Yes, with Manshi Rogeschwar, also a physicist, and they both have two children.' That was a lie, but it was fantastic." "You could see immediately that there was a gigantic potential on the one hand, but on the other hand you realized – oh, be careful."

What is real, what is a fake?

"And what new opportunities can we look forward to?" asked Illner. "A dam burst," Yogeshwar said. Development began in 2017, and last year it became apparent that these large language models had taken on a whole new quality. "We already knew Siri. But now there is an AI that is starting to be able to do everything at the same time: image recognition, text recognition and speech recognition. And one that is becoming more and more powerful and that optimizes itself." "And simulated the people," Illner added, showing the film of a deceptively real young Chinese singer who accompanies herself on the guitar, but behind whom there is actually AI software. #