Documentary 'More Human than Human' explores AI, by making AI

Have you ever thought about what it really means to be human in the age of intelligent machines?

More Human Than Human is a documentary that looks at the rise of artificial intelligence, partly from a technical point of view, but also from a cultural point of view.

As well as exploring the current state of AI, the filmmakers also wanted to immerse themselves in it. They took part in the building of a robotic AI to see if it could conduct the film's final interview.

More Human Than Human is running at Toronto's Hot Docs film festival. It was written and directed by Tommy Pallotta and Femke Wolting. Femke Wolting spoke to Nora about their film.

Here is part of their conversation:

So why did you decide to make this film?

We have always been interested in the crossover between technology and film, and of course there's so much debate now about AI; the last few years about how it's impacting the job market and basically more and more parts of our lives, in healthcare and education, in the economy and the financial markets. It's becoming part of almost every part of our lives without actually us really realizing.

And so we thought it was an important moment to make a film about AI and investigate from a human perspective what kind of changes AI brings to our lives.

So are are you worried about being replaced by an AI?

Yes and no - because on the one hand it doesn't live up to its promises right now. I feel that we are still in a clunky stage, but we also realized how fast computer technology is changing.

The film explored many different examples of AI. Which one amazed you or maybe frightened you the most?

A couple, on the one hand we felt like the examples with the people we filmed who are actually working with real robots. I mean the factories that we were filming that are robot-driven are kind of extraordinary, cinematic, and fascinating, but it's all still kind of repetitive simple tasks.

The AI that we felt was most frightening was actually the AI that's already being used today. It's much more invisible, it's in the internet, it's in the way data is used, and how so much about us is already known and used without us giving explicit permission, without any regulation.

Of course we're nowhere near being able to create the AI of popular culture: you know, the kind with true general intelligence that understands or has consciousness. What do you make of our desire to see human qualities in these machines?

I think it's part of human nature to want to build and develop new things and to take technology as far as it can get. And I think it also shows our need for human connection, so that we want to find these connections also in computers and in AI and that they can be meaningful.

For example, we filmed with these robot designers who had built these childlike robots and they placed them with elderly people in the Netherlands. And it was a big moral debate if they should be allowed to do that, if it was not frightening to have these robots. And within a couple of days these elderly people felt happier and they felt it had improved their lives.

And of course it's easy to say yes, human connection is better, and I think it probably would be better. But at the same time our society currently doesn't really provide that to a lot of people. And so to those people if these connections feel meaningful and it improves their lives. I think it's also a mistake to be really judgmental about it and say that it's not real.