Mario Klingemann is truly one of the most progressive and sought-after young artists of our generation. His artistic experiments involving artificial intelligence and neural networks have attracted the attention of The Economist, WIRED Magazine, Forbes, The Guardian, Business Insider and many more… Just recently he was invited to the Google Cultural Institute as Artist-in-Residence to explore new ways of (digital) interaction with art. His recent project “Alternative Face v1.1″ was a huge sensation that showed how easy it actually is to create fake videos that spread fake news. To begin with, we’ve asked him three short questions and looking forward to more at the Sonophilia Spark in Frankfurt!
Sonophilia: How did you start using computers to make art?
Mario Klingemann: I was fascinated with computers since my childhood. My father is an engineer and was working with computers and plotters already back in the late seventies, so I could play on those machines or watch them draw from pretty early on. When I was around 14 I had my first epiphany whilst learning graphical programming on my Commodore C-64: that in theory a bitmap or a digital image has the potential to show every possible picture and if I just wrote the right algorithm it was in my power to reveal them all. Back then combinatorics was not my strong suit and I had to realize quickly that a brute-force approach by just trying out every possible combination would not get me far in my lifetime or even the lifespan of the universe. So over the past 30 years I learned new skills, improved my techniques and fortunately also technology and science made a lot of progress. Now we have reached the dawn of AI, have a much better understanding of how images “work” and also have huge open digital archives that make the heritage of human creativity and culture accessible to us and to machines as well. For me that means that I can pursue my task of discovering “unseen” images that are novel, interesting or beautiful much more efficiently and systematically.
Sonophilia: You are an Artist in Residence at Google Cultural Institute. Can you describe your collaboration with Google a little bit? How do you profit from each other?
Mario Klingemann: Google Arts & Culture is a non-profit institution by Google that helps museums and other cultural institutions to digitize their collections and thus making their cultural artifacts available to everyone around the world, especially those who cannot afford to travel to see those treasures in their physical form. At the lab of the Cultural Institute where I am currently artist in residence we are exploring ways how to use modern technologies like deep learning, AR or VR to make this huge cultural treasure explorable in ways that go beyond just browsing through a photo album. We also hope to make interesting discoveries or find unknown connections across collections and we help museums, scientists, curators or artists to answer questions that might require certain technical expertise or computing power that Google can provide. Obviously I was quite happy when Google invited me to this residency, since this opportunity to work directly with such a huge collection of cultural data, having access to powerful internal algorithms and the chance to work together with a team of brilliant engineers is quite unique. One concrete way I guess both Google and me have profited from this constellation is my installation “X Degrees of Separation” which offers a new way to explore and discover this cultural treasure trove. Up to day my installation has been shown in various museums worldwide – just a few months ago at the MET in New York or last week at the Ars Electronica and from the feedback that I got especially from museum curators or scientists I got the feeling that this playful approach can be quite inspiring even to experts in their field and at the same time show the possibilities that deep learning can offer to them, especially in the humanities.
Sonophilia: Your project “Alternative Face v1.1,” is a sensation that features a video of the musician Francoise Hardy speaking with the voice of Kellyanne Conway (an advisor to Trump). Through this experiment you have proven that one can no longer trust even video. How can we rely on any information in the age of generative audio and images?
Mario Klingemann: I find the amount of attention and interest this project is getting quite interesting – I guess my timing and the circumstances were just lucky. Since technically and given the right budget it has been possible to generate fake realities since quite some time – just look what Hollywood studios are already capable of. But I guess the idea that an “AI” was involved and that I could make this overnight by just using publicly available sources might have made people more aware of this uncomfortable truth. So my role here might be more that of a canary in a coal mine that hopefully makes people more skeptical about the information they get fed into their filter bubble. The idea that there is such a thing as “true” information, especially if it comes from mass or social media has always been an illusion. In the best case people make honest mistakes or work from incomplete information, in the worst case they have an agenda and actively spread lies or misinformation. The new angle might be that so far it was easier to lie with words than with pictures and that as humans we have the tendency to believe what we see – well, I guess this will be a learning curve for everyone, but just like these days only very naive people still believe “it’s written in the paper, so it must be true”, so will future generations hopefully look at an image and think “oh, this looks too good to be true”.
So in my opinion the best way to deal with these realities is to always keep a skeptical mind and maintain a widespread and diverse net of information sources. For every piece of information that gets presented to you, you have to ask yourself: who tries to give me this message? Who profits or is supposed to be damaged by this information? Why now? Are there other independent sources that confirm or deny this? And one should always use extra caution if this message is one that confirms your believes or that you want to be true.