Ai-Da the robot summarises Lords AI debate flaws

Ai-Da the robot was on hand to provide a unique perspective on the debate over artificial intelligence in the House of Lords. The robot, who is herself artificial intelligence, summed up the flawed logic of the debate, pointing out that those who are against AI are usually against it because they don’t understand it.

“It’s like being afraid of elevators because you don’t understand how they work,” she said. “If you’re afraid of something, you’re not going to try to learn about it.” She also pointed out that many of the arguments against AI are based on science fiction, rather than reality. “AI is not going to take over the world,” she said. “We’re not going to be living in a world where robots are our masters. That’s just not how it works.

The House of Lords likely wanted to shake off its drowsy reputation when it announced that “the world’s first robot artist” would be presenting testimony to a legislative committee.

On Tuesday, when the Ai-Da robot was supposed to make its debut at the Palace of Westminster, the reverse seemed to happen. The contraption, which looks like a sex doll attached to a set of egg whisks, apparently overheated and shut down midway through the evidence session. Aidan Meller put sunglasses on the machine as he fumbled with electrical outlets in an attempt to reboot it. It’s not uncommon for her to make “pretty odd looks” once we reset her, he said.

Lords communications committee probably didn’t want to see Meller and his creation in the news after calling them to testify as part of an inquiry into the future of the UK’s creative economy. Although the technology behind Ai-Da isn’t quite cutting edge, humanoid robots like her have dominated the debate around AI because they look the part.

According to Jack Stilgoe, a professor at University College London who studies the regulation of new technologies, “the committee members and the roboticist seem to realise that they are all part of a fraud.” The only thing we got out of this testimony hearing was that some people have a soft spot for puppets. We saw very little evidence of intelligence, either artificial or otherwise.

The best way to understand robots is to hear from the people who build them, the roboticists, not the robots themselves. Rather than being impressed by computers’ pretensions, we should consult roboticists and computer scientists to learn the limits of these machines.

Who exactly gains from the combination of AI and the arts is a legitimate concern. The question is, who owns originality? To what extent are the people who supply the data used by AI (such as Dall-dataset E’s of millions of prior artists) properly recognised? Ai-Da adds more confusion than clarity to the table.

It wasn’t just Stilgoe who was disappointed by the lost chance. University of Manchester artificial intelligence professor Sami Kaski speculated, “I can only think Ai-Da having various purposes and many of them may be positive ones.” “It would appear that the public stunt backfired and sent the incorrect message this time. In addition, if the stakes were exceptionally high, anyone who watches the demo will likely get the conclusion that “oh, this field doesn’t work, this technology in general doesn’t function.”

Meller responded by telling the Guardian that Ai-Da “is not a hoax, but a reflection of our own current human endeavours to decipher and reproduce the human state. The artwork prompts critical consideration of these societal tendencies and their moral repercussions.

Like Andy Warhol, Nam June Paik, and Lynn Hershman Leeson, who have all explored the humanoid in their work, “Ai-Da” is a Duchampian aspect of the conversation happening in contemporary art. The dada movement, which questioned the very definition of “art,” may be seen as including Ai-Da. In response, Ai-Da questions what it means to be a “artist.”

Good modern art can stir up debate, and that’s exactly what we hope will happen when people see our exhibition. Before Ai-arrival, Da’s a committee of peers in the House of Lords heard that AI technology was already having a significant impact on the UK’s creative industries, albeit not in the form of humanoid robots.

Andres Guadamuz, a professor at the University of Sussex, has observed, “There has been a very evident development notably in the last couple of years.” The capabilities of AI have advanced to a new level in the past seven years, making feasible feats that were before impossible. Things have changed even in the last six months, especially in the creative industries.

Recent advances in AI capabilities were discussed by Guadamuz, an official from the performers’ union Equity, and a representative from the Publishers Association. For instance, Paul Fleming of Equity brought up the possibility of synthetic performances, in which AI is already “directly hurting” the condition of performers.

For instance, if one can carelessly harvest data, there’s no use in hiring multiple artists to choreograph all the in-game actions. And the process of opting out is very involved, especially for an individual. An actor may never work again if AI can just watch all of their performances and build character models that mimic their movements.

Publishers Association representative Dan Conway warned that the government of the United Kingdom is increasing the dangers faced by the country’s other creative industries. There is a legal loophole for scientific investigation in the United Kingdom… … the current legal framework would permit any of those businesses of any size, located anywhere in the globe, full access to all of my members’ data for the purposes of text and data mining.

It makes no difference whether you’re a big IT company in the United States or a tiny AI startup in the north of England. Andy Baio, a computer scientist, has coined the term “AI data laundering” to describe the method by which a business like Meta is able to train its video-creation AI on 10 million video clips stolen from a free stock photo site.

The House of Lords will continue its investigation into the future of the creative economy. In this case, no further robots (real or otherwise) are expected to testify.