Sci-Fi Bulletin: Exploring the Universes of SF, Fantasy, Horror and Spy-fi!

Part of Life Rewired, the Barbican’s 2019 season exploring what it means to be human when technology is changing everything, the new exhibition AI: More Than Human is an unprecedented survey of the creative and scientific developments in artificial intelligence, exploring the evolution of the relationship between humans and technology. Not long before the opening on May 16, Paul Simpson caught up with Assistant Curator Anna Holsgrove to discuss the exhibition and the role AI plays in society…

The exhibition sounds very intriguing. What’s your role – are you attached to the Barbican or are you just in for this?

I’m in for this as an assistant curator. I’m working with the two guest curators on the exhibition.

I became involved about a year ago when the exhibition was still in its development. I’ve worked on design and technology exhibitions before and because AI is such a fascinating subject, and feels so now more than ever, it’s really exciting to be able to work on something that’s new and hasn’t really been shown in this kind of exhibition format before.

One of the things that struck me immediately on reading the press release was that the two curators come from very different backgrounds.

Yes they do and I think that’s really shaped the show. Suzanne Livingston is based in the UK and she is an academic who specialises in cybernetics and philosophy. Maholo Uchido is based in Tokyo at the National Museum of Emerging Science and Innovation so the fusion of the two of them has really brought a unique vision to the show. It’s given it a real global perspective, which I think is really important.

Maholo has shown that the relationship with technology in the East is quite different from how it’s perceived in the West, and that’s a really important part of the show.

Although Suzanne and Maholo have different backgrounds, what they’ve wanted for the show has always been shared and they do bring these different philosophies to it. AI is such a global subject, it’s important that it resonates on that global scale.

Can you sum up those differences easily?

A lot of the Eastern perspectives comes from Shintoism, which is the belief that everything has a soul – humans, animal, matter, nature is all on this level playing field and that makes the relationship with machines that much more comfortable, whereas in the West there is the perception of robots taking over and things being that bit more scary, which isn’t necessarily true of AI, and the AI that’s surrounding us already.

So in a sense, the Japanese idea is coming more from a belief system as opposed to the West coming from literature…

You could say that, although there are equally beliefs from the West that we explore. The very first section of the exhibition, The Dream of AI, it looks at AI and also looks at the story of the Golem from Judaism – the story that in the 16th century Rabbi Loew brought the Golem to life and made the Golem, this character, out of clay and brought it to life to protect Prague.

The Golem is explored in fiction, things like The Incredible Hulk. It’s referenced through a lot of comics and films – in fact there’s an artwork in that first section, a film piece, that samples clips from a lot of these references: Blade Runner and things like the Hulk. The idea of bringing an intelligence to life is something that’s always been a desire and crosses a lot of cultures and dates back centuries.

That’s one of the key messages of the show – AI isn’t something that’s new and coming in however many years, it’s something that’s always been a dream and a desire and already a part of life.

In the West, discussions about robots are often a lot about control – Skynet controlling us, to use the obvious example – or us controlling the robots in an Asimovian way; is there an equivalent idea to that in the Eastern system, or is a totally level playing field?

I think it is much more of a level playing field, much more comfortable, a very different relationship and something I think that can be very positive. It’s a very good opportunity for visitors who’ve maybe not seen that side of AI before to encounter that as one of the first things when they step into the exhibition and leave with a slightly different outlook.

What’s the experience like for the visitor?

It’s a linear journey through the Curve gallery, but before you even get to the gallery, we’ve taken over the foyer as well so there’s a number of artworks that you encounter as soon as you step into the Barbican Centre.

Universal Everything have a piece at the Silk Street entrance, which is a machine learning piece. The visitor moves with this character which learns from them – it starts as a child-like form and learns from each visitor and become more agile, like a child growing. That’s something that people can engage with and have fun with before they go through the exhibition and then hopefully go back to it with more knowledge to understand what’s going on.

In terms of the narrative of the exhibition in the Curve, it’s split into four sections. The first, The Dream of AI, is where we look at these belief systems and AI dating back centuries, and this desire and this want to create beings from other forms.

The second section, Mind Machines, is where the beliefs become more tangible and we look at people like Alan Turing, Claude Shannon and important breakthroughs in AI, leading up to Move 37 and AlphaGo. There’s an interactive exhibit looking at how the machines beat a human in AlphaGo and also what that means – this idea of using games as a testing ground for AI has much wider applications.

The third section, Data Worlds, is where we look at those applications in more detail so this is where we really show that AI is already very much entwined with our lives in things we’re already doing, whether they’re things we expect or things that are more surprising. There’s a piece by Nexus called Micro to Macro which shows how AI is being used at those different scales – classifying things, such as in cancer diagnosis looking at energy in a way that humans aren’t able to do as effectively, and then on a bigger scale. That’s also where we look at a couple of the ethical questions that arise with AI, so looking at bias and privacy.

The fourth section, Endless Evolution, is where we speculate and look a bit further ahead so that’s when we have a film clip from Ex Machina and we look at how we might evolve as a species. We also have a film, Visionaries, where we’ve interviewed six people about what they think the future might be about, and how it will change.

How much are you using pre-existing material, and how much has been commissioned specifically?

That’s been one of the real challenges because the goalposts kept shifting as well when we were curating it. We have got a lot of new commissioned artwork and in the foyer we have new commissioned artworks. Mind Machines tends to be more historical content, but there’s a new digital timeline which people can dive into and explore in more detail. Data Worlds is almost entirely new interactive works where we’ve been working with the artists and the media designers almost till the last minute to see everything is relevant and up to date.

With AI of whatever sort, it’s something that’s created by humanity, so do you deal with the issue of computers programming themselves?

We look at how humans influence what’s going to happen, and the main topic we look at that through is bias, which is a huge problem, and something we’re very honest about.

If there isn’t diversity in datasets that we’re training machines on now and continue to train machines on, they’ll pick up human bias and human flaws, so in that sense we’re saying that it’s not the machines and the technology that we need to be wary of, but what’s being put into them, and that’s where we need to tread a bit more carefully. On the whole the exhibition takes a positive approach – we want to show people the opportunities that AI has because we’re aware that people do often find it daunting.

In movies, AI is usually shown as non-beneficial – there’s ET for friendly aliens but friendly computers not so much; the drama comes when the AI goes wrong.

That’s why we have been quite selective in the films that we’re showing and the clips from them that we’ve chosen to use because although AI in films is a lot of people’s first encounter and features heavily as a story, we do want to show that AI is beyond the realm of science ficton and is already part of day to day life and does not need to be feared.

In your time working on this, what element has surprised you personally most to discover?

I think learning from Maholo about the Eastern philosophy has been a big eyeopener, and discovering how much AI is already part of the systems that we use: things that can be quite mundane, like banking and how traffic is controlled, to much more revolutionary things like AI being used to detect eye disease or cancers much earlier and used in this collaborative way. People might come to the exhibition with the view that it’s AI or humans and we’re really keen to show that it’s this collaboration between the two which produces something that’s very interesting and exciting.

AI is a tool and any tool is only as effective as the person using it.

Exactly.

What’s been the most exciting part of the project for you?

Working with the artists, delivering the new commissions – as you just said, AI is another tool, and it’s another tool that artists are using in their palettes. Seeing artists bringing pieces to life, like the Universal Everything piece.

We’re working with Lawrence Lek on a really exciting piece that looks to 2065, and is a video game that is specific to the Barbican. He’s modelled the Barbican in 2065 and you can explore this future world and see what life will be like then.

AI: More Than Human runs 16 May–26 Aug 2019

Mon-Fri £15, Sat , Sun & Bank Holidays £17(Concs/Students/14-17/Art Fund £10-£15); Young Barbican £10, Under 14s go free