Refik Anadol: “I’m imagining a different world”

Refik Anadol: “I’m imagining a different world”


Machine Hallucinations

Mr. Anadol, do you believe that art should be for everyone?

Absolutely, and the funny thing is that when I say that art should be for anyone, everyone, at any age, in any culture, I find it fascinating that that is a very different approach than many artists. But I’ve always been a bit of an outlier, because when I started making art, I didn't have a classical training, I wasn’t trying to make a gallery show or any of those predictable pipelines of art. What I was most inspired by is making public art, meaning art outside of places designed for art, so no museums or galleries. I found much more excitement, motivation, and connection with people when presenting public art — that can be transforming a building into a canvas: a school, an airport, you know, places that are not necessarily designed for art, but places where art appears unexpectedly.

It’s even more special to find your art in unusual spaces, like when you mapped your work, Living Architecture, onto the Gaudi Museum in Barcelona. Art becomes much more accessible in that way, and the whole city came out to see it.

Right, because if I’m going to see art in a museum or gallery, that’s an intention, and for me, that’s not enough. Or at least, that’s just one way of doing it. There should be more ways of doing art.

“With my work, personally, I try to find things that connect us, things that belong to humanity. This is where I found a whole new connection, a new form of storytelling, a new language.”

Even though it wasn’t part of your initial goal, you do show work in museums and galleries now, although it seems like you’re careful to choose environments that work in tandem with the kind of art you make.

Yes, I am in love with architecture as a canvas, my work is always in conversation with architecture. I do not have any work that has no connection with the architecture. When I see the Gund Lobby of the MoMA, which is an incredible space where I showed my recent work Unsupervised, for example, it is a place we can turn into public art which was never used before in such an eclectic way. We’re transforming 200 years of art from the MoMA and putting it on the wall, and incorporating the changes in the surrounding environment, the movement in the room, the light, the weather, all of that affects the imagery. So that's a very different way of making and showing art. With my work, personally, I try to find things that connect us, things that belong to humanity. This is where I found a whole new connection, a new form of storytelling, a new language.

You’re talking about the kind of public-use AI-powered language you use to craft your artwork?

Well, the language can be abstracted language, it can be any vocabulary; between machine and human, brain signals, heartbeat data, Bluetooth signals… But I’ve found that human and environments and machines; that triangle is our future. There is no more future without machines, unless we unplug ourselves! (Laughs) I’m not solely existing in the virtual world with computers and algorithms. I’m not! I love the physical world too, but I do find my language in the virtual. With this language, I'm imagining a different world.

How do you reconcile that technology-focused approach with the kind of human or emotional touch you emphasized earlier?

Well, the kind of language I’m using, wherein I source public-use images from social media and the Internet around the world and feed it into the algorithms of AI, that breaks down our cultural barriers. It uses the collective memories of humanity to cross space, time, culture… It’s a very utopian dream.

True. Plus, even at its core, AI art still has a human being working with it, making decisions and conceptualizing ideas.

That’s something that’s very important because when people think about AI and art, they imagine a button that the artist clicks and then an art pops up or something like that. It is absolutely the opposite. It takes significant time. For example, because I live for years now in LA, where we have a strong connection to nature, I decided to make an artwork dedicated to the California landscapes which have inspired me the last 10 years. So I started by thinking about the places to include, then I started downloading the data of these landscapes from social media around the world. Let’s say there are 155 million images of the landscapes. It took six months to make sure we only get two million images which just show the landscape — nothing private, nothing personal, nothing human, just the landscapes. This sentence is another giant research. And this is not just one click!

It’s a process in itself…

Yes, it’s training AI to be sure that it only gives back what I'm looking for. It's like an incredibly deep research. And once done, that’s when the pigmentation of AI starts, which is another couple of months. To make AI  dream in the way that I want it to, that is a process that is computationally heavy, but also very hands-on, there are parameters that I'm choosing, there are moments that are good or not. It is an absolutely an incredible amount of work to the point that it feels like a dialogue or a collaboration is happening. And once I feel like this AI can be the one that can dream these specific landscapes, on top of it is my own artistic composition, is my signature style, the shape shifting and moving and colors and speeds and the behaviors. AI art is not that you type something, and something pops up. That's not art. That is like a tool that just creates a bunch of images or a video or text or a sign. Mine is a process that involves data training, AI and artistic computation — and all together that is my process. That is what machine hallucination means.

When did your process shift from using existing images, to producing your own images that you feed the algorithm with?

That idea started most likely in the pandemic. I had this very heavy hit; I remember the second week of the pandemic here in California, during the initial shutdown, we couldn’t go to nature… That was a major realization to me! We can’t go to nature, so maybe nature can come to us. That’s when I started data collecting, not only online but in the physical world around me, and I said to myself, “Okay, when things open up, I’m doing it.” And that’s when I was working on my Glacier Dreams project, which uses publicly available data and archives together with images collected by me in Iceland, and processed through machine learning algorithms and transformed into AI-based narratives.

It’s like you said before, although you work in the virtual world, you have to remain grounded in the physical world, too.

Absolutely, I mean even now I’m living in Amazonia, and being in the rainforest is very different than just talking about the data. This connects me to the topic in a very true and sincere way. Opening the narrative can only happen if the artist is in charge of data, otherwise, it's just another tool that doesn't have this special perspective of an artist. That's my learning.

Thanks for sharing!

Simona B.

Founder/CEO at AvidaLAB || Marketing and Branding Strategist || Art Historian and Art Dealer || AI Enthusiast || Helping Artists achieve their dream goal

1mo

One of my favorite artworks by Refik Anadol, The intervention on the facade of Gaudi's 'Casa Batlo''

  • No alternative text description for this image
Caterina Ciccotti

Senior Editor | Consulente Editoriale | Book Pusher | Translator | Localizer | Transcreator

1mo

Architecture as a canvas!! Human and environments and machines as the triangle of our future - our language! Thank You, Simona B., for these necessary, perfectly chosen words by Refik Anadol.

To view or add a comment, sign in

Insights from the community

Others also viewed

Explore topics