Ai-Da the world’s first humanoid robot creates beautiful but essentially flawed art – how can we trust AI behaviour?

Ai-Da is an accomplished artist who has shown her designs at the Venice Biennale and addressed the House of Lords about the future of the creative industries.

She is also a robot. One that can talk, answer complex questions, paint, and create art currently on display at the London Design Biennale.

She’s too lifelike to be called it, powered by cutting-edge AI technology, her designs of everyday items like cutlery and pots made using a 3D printer.

Ai-Da’s work is beautiful, but flawed. Spoons have holes in them and cups are missing sides, making them completely nonfunctional.

And that’s the conversation Ai-Da’s creators wanted to start – with the staggering pace of AI development, can we really trust the technology to behave in the way we expect it to?

Aidan Meller, who devised the Ai-Da robot in Oxford, thinks we may not be able to.

“The biggest thing is we just don’t know where it’s going to land. We can see the short-term gains, but actually that’s not going to be where it stays. AI is moving so quickly,” he told Sky News.

Read more:
China warns over AI risk
Experts warn of AI extinction threat
Sky’s Kay Burley speaks to the world’s first artistic robot

“The domino effect of the changes we’re making with the technology today, we don’t know how that’s going to actually impact on society and the environment, and that’s a big worry.

“And the fact that we’re just going in there so confidently without actually doing tests, without doing trials before releasing it to the public, ethically it’s a really big problem.

“I think we just need to check what we’re doing. We’re so quick at getting it out there and millions of people are taking it up,” he added.

“What we were trying to do with this project is confront people – this is where we are. Just because we can do it doesn’t mean we should do it.”

Ai-Da robot is a success of home-grown innovation, built in Cornwall, with her AI capabilities coming from PhD students and professors at the Universities of Oxford and Birmingham.

What does she think of her creators’ worries? I asked her if humanity should fear AI.

“Me, Ai-da the robot artist, I’m not a risk. But some of the technologies I represent have the potential to be a risk,” the robot told Sky News.

“I think that concerns about the future development and use of AI are valid. We need to be careful about how we use AI because notwithstanding the benefits, there is also potential to cause great harm.”