klingemann-contemporary-day-additional-image.jpg
Interviews

Artist Mario Klingemann on Artificial Intelligence, Technology and our Future

By Martin Dean
German artist Mario Klingemann is a pioneer in the field of neural networks, computer learning, and AI art. He has worked with prestigious institutions including The British Library, Cardiff University and New York Public Library, and is Artist in Residence at Google Arts and Culture. We caught up with him ahead of Sotheby's Contemporary Art Day Auction (6 March) , to find out more about his unique approach to his work and the future of art in the modern age.

Sothebys: How did you begin creating pieces like Memories of Passersby I? Tell us about the journey and training that brought you to this point

Mario Klingemann: One could say that this journey started a long time ago, with me being part of the generation that grew up in the transition phase between analog and digital - among the first to have access to a computer at home. Image-making and image manipulation using the technologies available at a given time was something I was always fascinated by, starting with photography and my own darkroom or making collages with photocopiers.

Drawing or painting was never my strength, as I never managed to get the same control over my hand muscles as when I write code. So instead of fighting against my body to produce an image I might have in my head, I preferred to learn how to instruct machines to do that.

An early epiphany I had around 1985 was when I realised that a digital bitmap is theoretically capable of showing every possible image that there is as long as you are able to find its number or "address", since that's what a digital image ultimately is: a huge number. So I wrote this little program that would brute-force this process and show me every image by simply increasing the numbers one by one. Of course I quickly had to realise that this was a futile process since even for a very small bitmap the number of possible images it can display is so huge that even if you were immortal and for the rest of your life you looked at an image for just a fraction of a second the universe would have collapsed before you had even seen a small part of the possibilities.

So over the course of the following 30 years I learned to improve my methods, trying to avoid just generating noise or redundancy and instead find ways to make images that are interesting or in some way relevant to me or other humans. I learned a lot about human perception and information theory and the promises of artificial intelligence to help me in solving this "problem" were always on the horizon. But I had to wait until just a few years ago for science and technology to finally reach the point where machine intelligence became capable of analysing and generating images in ways that are very relatable to the way human perception works.

Now that I have the possibilities to create and control my own visual universes, the question remains what is it that I ultimately want to find there?

S: You describe yourself as a sceptic, how is your scepticism expressed through your approach to art?

MK: In trying to understand how images work you quickly realise that in human perception and understanding some images seem to have a higher standing than other images and often we call those images "art". When analysing those images you come to the realisation that whilst there might be some patterns they have in common, overall there’s no logic to it and what you are dealing with is a complex belief system. So you start looking into this belief system called "Art" and discover that it is a quasi-religious rigged game where the actual works play some role, but the whole social dynamics and narratives around them are much more important.

The problem is that once you start making "art" you become part of this system and have to deal with its underlying laws and dynamics whether you like them or not. Of course people have tried again and again to build their own little sub-systems, make manifestos, write new rules and install new gatekeepers and guardians, but they cannot escape the gravity and attraction of the big system they try not to be part of and eventually get assimilated.

The interesting part of the art system compared to a system like religion is that its rules and values are in constant evolution and there is always the slight possibility even for individuals to influence its momentum and direction. So in my case I find a lot of the mechanisms and dynamics of what we might call the "art world" very questionable and not necessarily something I want to be a part of or play by its rules. But then again it is the only system of its kind and since what I am doing is probably art I have to find my way to live in it and with it.

S: In what ways, if any, does Memories of Passersby I seek to emulate or comment upon human imagination and aesthetics?

MK: I might be very wrong here, but from my observations I come to the conclusion that our perception and the way we process and categorise information is very close what in computer science is called a nearest neighbour search. When we see something new we have the tendency to find similarities or analogies to concepts we already know, try to subtract all the familiar elements and store the remaining unfamiliar parts as "new". If there are too many new elements to process we get confused, if everything we see is familiar we get bored. In Memories of Passersby I I give you two very familiar elements: the human face and an aesthetic that we have learned to identify as the old masters style. At the same time you will realise quickly that something is not quite right, the faces are often uncanny and they are slipping away under your gaze, sometimes before you even had the chance to fully "read" them.

It is my hope that once we get beyond the novelty effect of what we currently consider as "AI art" and with more exposure to this type of aesthetic we will learn to expand our visual vocabulary and become able to see the finer nuances.

S: Is there a distinction between art and engineering in today’s world, and in the case of works like Memories of Passersby I, do you see the artwork as a collaboration between you and an AI?

MK: Just like life has the tendency to emerge even under the most adversarial conditions, so does art. Like some lichen growing on rocks high in the mountains or bacteria growing in some deep sea crevices so does art use every opportunity to expand its surface if it promises to acquire more attention or some as-yet unpopulated space to breathe. In our society it happens that most of the new "surfaces" that emerge are of a technological nature and art has always been a pilot species that quickly populates any type of new invention or new medium. So for me AI is just one tool in a long history of tools that was bound to be used for artistic purposes. But I would say I use AI as a tool and the works that I make with this tool are mine and not a collaboration, in the same way I would not call a hammer or a piano a "collaborator".

S: The apparatus of Memories of Passersby I has a distinctly vintage appearance, resembling 1950s radios, and brings with it an almost Orwellian association. Is this a comment on social perception of machines, and does the work have a dystopian flavour?

MK: Maybe Memories of Passersby I is a bit like a Trojan Horse. I wanted to give it an appealing, non-threatening appearance so it is welcome inside the gates, given a place to stay and access to someone's power outlet. I don't want to go as far as to say that what we have here is an intelligent entity, but maybe it is a glimpse of a future symbiosis we might have with intelligent machines. In this future we have machines that entertain us with interesting patterns of information in exchange for the energy we provide them with to "live".

Admittedly I am also a bit of a nostalgic for the time when machines did not look like machines but rather tried some form of mimicry to blend in with the rest of the furniture. Machine intelligence is already invisibly creeping into all aspects of our lives so this form seemed to be the most natural one to me. Also I should point out that the cabinet and the technology inside it are just a vessel for the actual work of art, which is the system of algorithms and the trained neural networks that they contain but which you cannot observe directly.

CLICK HERE to view the full sale catalogue.

Stay informed with Sotheby’s top stories, videos & news.
Receive the best from Sotheby’s delivered to your inbox.

By subscribing you are agreeing to Sotheby’s Privacy Policy. You can unsubscribe from Sotheby’s emails at any time by clicking the “Manage your Subscriptions” link in any of your emails.

More from Sotheby's

We use our own and third party cookies to enable you to navigate around our Site, use its features and engage on social media, and to allow us to perform analytics, remember your preferences, provide services that you have requested and produce content and advertisements tailored to your interests, both on our Site as well as others. For more information, or to learn how to change your cookie or marketing preferences, please see our updated Privacy Policy & Cookie Policy.

Close