Instead of panicking, let's learn from the machines

For more than 90 years, people have been processing their fascination with artificial intelligence, but also the hopes and fears that it arouses in them, also in films. The likely first feature-length AI flick is Metropolis. Fritz Lang's vision of the future from 1927.

Shot in the turmoil of the Weimar Republic, the film depicts a gloomy world with a strictly divided two-class society: the rich live in the upper town, the workers toil underground on huge machines that set the pace.

The German silent film classic remains highly topical, because the fear of the power of machines is still present: the image "Théâtre D'Opéra Spatial", created with the help of the Midjourney AI system, recently won an art competition in the US state of Colorado and thereby sparked a debate about the nature of art.

A user named OmniMorpho tweeted: "We're seeing art dying right before our eyes - if creative professions aren't safe from machines, then even highly skilled jobs are at risk of becoming obsolete. What will we have then?” (Tweet translated from English using AI).

OmniMorpho paints the picture of a world dominated by AI, in which the creative human being is degraded to the servant of an intelligent machine. Yes, if that were the case, it would be truly creepy.

Of course, machines are not inherently evil. They are the product of us humans. This is already shown in "Metropolis": While Freder, the son of the sole ruler Johann Fredersen, falls in love with Maria from the lower town and changes sides, the inventor Rotwang creates a human-machine that cannot be distinguished from the real Maria. With her he wants to mislead the oppressed workers and seize power in Metropolis.

The "Théâtre D'Opéra Spatial", the "Space Opera Theatre", is also the product of a human being. And it was work. You feed text-to-image generators like Midjourney or Dall-E with descriptions of what you want to see, and the AI ​​makes a selection. If nothing suitable is found, you have to rework it or start a new attempt.

"I came up with the text input, tweaked it for many weeks, and curated all the images that were produced," says Jason Allen, head of Incarnate Games, who submitted "Théâtre D'Opéra Spatial" to the competition.

Covers, posters, t-shirts or websites — yes, AI makes it easy to illustrate in almost any style! However, designers will not become superfluous. Just like my favorite Italian around the corner hasn't closed just because I can cook my favorite pasta at home. Behind every good work there is (still) a creative person.

Blanket rejections or vague fears of technology are neither beneficial nor justified. The sprinter Gina Lückenkemper raced to the EM gold over 100 meters this year, and all of Germany cheered, although everyone in their car is faster over the 100 meters.

For the former world chess champion Garry Kasparov, his defeat in 1996 against the chess computer Deep Blue was painful, but ultimately, as he emphasizes, a blessing: “We thought we were unbeatable in chess, in Go, in Shogi. All of these games have been gradually pushed aside by increasingly powerful AI programs. But that doesn't mean life is over. We need to figure out how to use this to our advantage.”

So we shouldn't panic when AI systems are sometimes superior to us, but we should be very attentive to the genesis of this key technology so that knowledge and assessment are spread over many shoulders instead of just being left to those who develop AI. Violence, child pornography and vulgar and compromising content are of course reprehensible, whether created together with AI or not.

And fortunately, abuse is curbed by the fact that EU law allows an author to object to the commercial use of their works to train machines.

AI reflects admiration for the most intelligent machine we know — humans. Medicine Nobel laureate Francis Crick saw us as nothing more than a bunch of neurons. For me, intelligent behavior is nothing more than a bunch of algorithms: what are the recipes for rational behavior or human reason?

Finding your recipes doesn't mean our lives are over. Every innovation challenges us to figure out how to use it to our advantage.

Kristian Kersting is a professor for AI and machine learning at the TU Darmstadt, co-director of the Hessian Center for AI and winner of the "German AI Prize 2019".

"Everything on shares" is the daily stock exchange shot from the WELT business editorial team. Every morning from 7 a.m. with our financial journalists. For stock market experts and beginners. Subscribe to the podcast on Spotify, Apple Podcast, Amazon Music and Deezer. Or directly via RSS feed.