PATRICK TRESSET: ROBOTS AS STYLIZED ACTORS

Patrick Tresset: There are many reasons why I use robots. On the one hand, robots enable me to explore the drawing practice with a conceptual and emotional distance. On the other hand, I use them as actors in performative installations.

I prefer to say that I use robots rather than collaborate with them. In my view, for there to be a real collaboration, they would need to be aware. I have an experiment that gives this impression, where the robot continues a drawing I began. It's interesting as a research outcome, but I can't see how it could work as an art installation.

PT: I design, develop, and build robots, and I write the computational systems that control their behavior. I use them as actors in several performative installations where they make marks, most often on paper. I also use them in the studio to produce drawing and painting series. But why? It just happened, and it seems to fit my abilities and desires. It is easy for me to express myself in this way.

PT: I was a painter, or at least trying to be. Initially, when I began using computers, it was to explore drawing and painting practices with emotional, psychological, and conceptual distance. In the late 1990s, for several reasons, I faced a severe artistic block; more than a block, I was at a dead end. I had always kept an eye on the evolution of computers, and I knew I could code, as I had learned when I was a kid. I thought I could do something with them, and the switch was quite painless. I felt very comfortable with this new medium, and surprisingly, I had fewer doubts.

Around 2000, the Internet allowed me to learn about the algorists, notably Frieder Nake, Roman Verostko, and, of course, Harold Cohen, and I began to use pen plotters. At the beginning, I was salvaging discarded office computers from skips, installing Linux on them, and I also came across the programming language Python. I got some interesting results relatively quickly and was already producing works depicting faces. These first drawings constituted the portfolio that enabled me to get into a new master's program at Goldsmiths College in London in the computing department. By then, I had realized that I needed to develop better programming skills and a more scientific approach.

The robotic elements came later, during my doctoral studies in 2009. It became evident that to explore the drawing practice, I needed the physical motion of a pen moving across a surface to embed its movement’s history in the sketch. As soon as I started working with robots, I found an interesting space to explore.

Of course, I wanted to continue investigating mark-making, but from the first time I exhibited at the Kinetica Art Fair in London in 2010, I discovered something I hadn't expected: robots are natural actors. Using them, I could play with humans "artistically" in performative installations. That realization opened up a new practice, and it took me a couple of years to appreciate the possibilities.

This combination of research, drawing, performances, machines, and computation was most suited to my capacities, allowing me to express myself most fluidly and naturally. I need this interface to express myself. Everything has to go through computational systems, often embodied ones. The reason is that I can’t stand seeing traces of my own spontaneity.

PT: I can’t generalize, but of course, I can only tell you about my own work. When I was a painter, I loved to sketch very spontaneous drawings, but as a human, it is very difficult; sooner or later, you realize what you are doing, and then it is not spontaneous anymore.

Using robots resolved this issue, as I don't even touch the pen, and my influence is prior to the execution since I design and write the systems. In a certain way, a part of myself is embedded in the code. My original doctoral research aimed to find out how a system could produce drawings that act in the same way a human-made one does in terms of perception, without being pastiches.

In a certain way, my early robots and the ones I exhibit most often are "naive" drawers, as they don't know what they are doing. The execution is not contaminated by their knowledge. They are reactive systems based on robotic AI theories from the late 1980s.

I have other systems that use machine learning, so the drawing is influenced by their "knowledge," but there is still a lack of humanness—an absence of intention—which is what I was looking for in the first place. Now, I am interested in exploring the possibility of bringing back a human presence into the drawing process.

One important element when using robotics is embodiment and time; robots act in the same physical space as us, with the same rules. For several reasons, I generally don’t use industrial robots but rather custom-made ones. The former are designed to be as precise and rigid as possible. If you use them for drawings, the output is the same as using a pen plotter; they produce lines with no memory embedded. The ones I design, which are dedicated to drawing, have a little flex, which means that the characteristics of the motion that created the line affect its appearance. When you look at a drawing, you can see if it was sketched rapidly or gently. In the installations, I often have multiple robots, and I tune them differently—one "nervous," one softer, etc. The former is identified as being a bit rough and mad, and its drawings are interpreted in the same manner.

There is an interesting research paper about this by Alessandro Pignocchi: HOW THE INTENTIONS OF THE DRAFTSMAN SHAPE PERCEPTION OF A DRAWING. Similar to a human-made drawing that has its history embedded, which affects our perception of the drawing, the same is true for sketches produced by my robots. The drawings' appearance is not forced; they are "true" to the robots. The style is a consequence of the robot’s physical characteristics, its perception, its motor abilities, including the bugs.

PT: I can tell you about those who take part in my HUMAN STUDY #1 performances, where several robots draw a human in a 20 to 30-minute live session. It is conceived to provide the sitter with a rich, multilayered experience and to be appreciated as a performance by the audience. More people will experience the piece as spectators than as performers.

People often mention how the robots look at them, and their movements are interpreted as betraying certain personality traits. All of this is due to the behavioral elements that I added and designed to affect both the sitters and the audience. This is why I refer to my robots as stylized actors and the installations as theatrical. The robots function as actors in the same way that puppets do, with the added characteristic that they are, to some extent, autonomous and perceived as such.

Designing and planning the scenography enables the installation to be an artwork, an "oeuvre," just as there is a distinction between reality and films, theater, or literature. It must work artistically when exhibited in museums or galleries within a contemporary art context. This is challenging; due to the nature of the piece, it can easily veer toward entertainment. Generally, having people sit perfectly still for the duration of the session, along with the mise en scène, makes the piece work in a contemporary art context.

When I am present for the opening of an exhibition, I retune each robot until I am satisfied with the drawings. I also always sit for the first performance, staying still for the duration of the session; this encourages others to try to remain as still as possible, effectively making them performers. I even started projecting a video of a performance to help people understand what they need to do. This is essential; without it, the installation doesn’t look right, the performance quality suffers, and the sitter's experience is not what it should be.

I feel privileged to have this piece, which I have exhibited for thirteen years. I am continually working on it, and there are many different versions. I still take great pleasure in exhibiting it, and it continues to feel fresh for audiences.

I keep all the unsold drawings as part of an artwork titled "Collection," which already includes more than fifty thousand works, each serving as a memory of a performance with an individual.

PT: Time passes quickly. I have dedicated as much time as possible to my art practice for three decades. I started as a painter in the 1990s and then began using coding before 2000. Anecdotally, I made my first animation in assembly code in 1977, followed by algorithmic images in BASIC in 1981 and then Pascal. While I wouldn’t call it art, it was computational and visual.

To return to your question, I was only able to focus exclusively on my art practice from 2009 onwards, without needing non-art-related, bill-paying occupations.

From my perspective, the artist's role has not changed with the evolution of computational technologies. More precisely, the questions you must ask yourself and the problems you need to resolve as an artist remain the same: what? why? how? (perhaps not in that order).

I believe one of the roles of art is to provide a function and a place for artists in society; of course, there are other roles. This is how I see it: I have found my place in society as an artist. That’s not all I could do, but it’s the best I can do. It is my role, which has nothing to do with technological evolution.

Over time, new mediums have been added to the repertoire, and AI is one of them. They allow artists to express themselves differently and develop new practices and languages. In theory, a new medium enables the development of a personal language, as there is no established academism yet; this was true with new media when I joined. It should also be the case with art produced using AI systems, but surprisingly, it doesn’t seem to be.

Contemporary AI offers new possibilities for producing images, videos, and now stories. However, image-making is sometimes confused with art. The various movements and explorations of the 20th century have demonstrated that any medium or technique can be used to create works of art.

I have been experimenting with machine learning and AI since 2008, but I rarely mention it. I’m not sure how to express this, but for me, AI is nothing special—just a different computational paradigm, albeit a fascinating new technology. I wish that the fact I use AI and robotics didn’t affect the way my work is perceived.

The ways art is appreciated, consumed, disseminated, and acquired have changed over the past 10 to 20 years. AI will further modify the population’s needs and desires. Our tastes have already become more impulsive due to recommendation algorithms and the importance of the attention economy. It may be pessimistic to think that AI will enable mass manipulation far more efficiently than before, but it is certainly already the case.

PT: In 1991, I moved from Paris to London and decided to become an artist. From then on, this would be the focus of my activities. It now seems like a normal decision.

My mum attended the Beaux-Arts school, and I also had an uncle who was an architect but spent all his free time on his art practice. He was, I think, the most influential figure, as he showed that you could naturally be an artist. You could fully focus on some ideas and have a practice—a vocation.

My dad was an engineer in charge of automating the foundry where he worked. During this process, he attended a seminar and brought back a computer that was only programmable with assembly code. There wasn’t even a screen, but it was enough to fascinate me with the potential of creating autonomous systems. Thinking back, I also saw videos of industrial robots in 1976; this might have influenced my imagination.

So, the idea of being an artist came from childhood, and if my ten-year-old self could see what I am doing now, he would be very excited, though perhaps not surprised.

PT: I was lucky to be there just at the beginning of the Art and Computational Creativity program when the purely Math and Computing department was being transformed into an Art and Computation one, more in line with Goldsmiths' reputation in contemporary art.

I joined as a mature part-time student, and from the beginning, during the MSc, I worked well with Prof. Fol Leymarie, the course director. After the master's, we decided that I should pursue a PhD. The first year was dedicated to finding funding; with Frederic, we produced a good proposal and secured significant funding for three years, which gave me great freedom. We co-wrote papers, presented at conferences, and participated in summer schools, etc. I even taught a course in creative robotics for a couple of years.

Being in the academic world at that time meant I met pioneers such as Frieder Nake, Roman Verostko, Harold Cohen, Paul Brown, William Latham, and others. I was also in touch with the Computer Arts Society in London then. This context has undoubtedly been influential.

During my studies at Goldsmiths, I learned to be more rigorous in technological research, AI, and robotics, and to have clear artistic intentions. I also learned to give lectures and talks and to communicate effectively, which has been very important in my career development. However, this skill seems to have become quite rusty now, and I need to practice it a bit more.

Goldsmiths is a great place; it offers a lot of freedom. As long as you do something interesting, nobody will stand in your way, and you will also receive some support. It is also beneficial to have the Goldsmiths "stamp," as when there are high piles of submissions, perhaps yours is read with more attention. The same goes for funding; at the beginning, it helps to have these kinds of credentials.

Another influential aspect of my time at Goldsmiths was the contact with students, researchers, and ideas from other disciplines, such as anthropology, speculative design, cultural studies, and psychology.

PT: I became aware of crypto ideas and blockchain fairly early on and was familiar with artists working in this direction. When I was at Goldsmiths, I attended a workshop where Rhea Myers was participating. To be frank, I don't recall much about it, and looking at the dates, it’s likely that she spoke about crypto, but perhaps I am mistaken. I also visited some Furtherfield exhibitions where the subject was present.

My attention was particularly focused when Robbie Barrat got involved, and I was also in touch with Serena Tabacci, though I can’t recall how.

I wanted to join in, but I felt the need to produce something new—something adapted to online appreciation and suited to the NFT space and its mood—when all my practice was physical. I was very busy with exhibitions in 2018 and 2019, and even during the first year of the pandemic, I couldn’t find the time.

In September 2020, Memo Atken, whom I know from Goldsmiths College, began publishing his findings on the environmental cost of crypto and NFTs. That halted things for me. Later, in early 2021, Joanie Lemercier led two or three meetings with other artists to introduce us to NFTs, while everyone was simultaneously searching for environmentally friendly solutions. Mario Klingleman participated in some of these meetings as well. Then, in the first week of March 2021, he minted on hic et nunc, and many artists, including myself, followed suit; I minted my first objkt on the 5th.

For all the artists who were at the beginning of hic et nunc, those first few months are somewhat legendary—something amazing. It was about realizing that there were so many artists from around the world, both from the global south and the north, exchanging ideas and developing something together. These times really reminded me of when artists occupy a dilapidated part of a city. In the beginning, there is the same energy: parties, collaborations, enthusiasm, etc. It was the same but on a global scale. It allowed artists from the global south to earn incomes to live on during a time when freelance opportunities were severely reduced because of the lockdowns.

During the second year of the pandemic, opportunities for exhibitions were extremely limited. Apart from providing me with the income to pursue my work and pay for my studio, after hic et nunc’s emergence, I rapidly developed a computational system that enabled a digital-based output, giving me a new means of expression and a new area for my practice.

Another interesting adventure linked to NFTs is alterHEN. After a few months, when the atmosphere on hic et nunc became less artistic, Diane Drubay and I launched a gallery—an NFT platform based on Tezos, alterHEN (alterhen.art). With a group of twenty selected artists and funding from the Immaterial Future Association (now Culttech), we promoted ideas on the potential of NFTs, as we saw them as having the capacity to become a new cultural activity.

One of the most interesting elements of NFTs is the global community that has developed around them, including collectors, artists, curators, and more.

PT: No, not at all. I was attracted to the energy and youthfulness, but also, as an artist, having income streams is very important.

PT: For the installations, it takes me years to go from the first idea and concepts to sketches and implementation, but each installation is, in practice, a series. I am still working on ideas I had in the early years of exhibiting these types of installations.

I generally spend a lot of time—months, in fact—conducting computational research on sophisticated elements for the installations. However, when I begin testing and working on the artistic side of the implemented installation, I remove most of these sophisticated elements.

When I premiere a piece, it typically takes me a couple of weeks to refine it into something that works artistically. This is due to the performative aspect, which necessitates a public rehearsal period.

For digital works, at the beginning of hic et nunc, I developed generative software with a graphical interface that is very versatile and enables me to produce multiplayer animations or still images. From concept to final work, this process can be completed in a couple of weeks; of course, sometimes it takes longer, but never more than a month. I enjoy this possibility to be spontaneous, as it adds something interesting to my practice.

For the painting series, everything is planned and well-defined before I begin production.

PT: My interest lies in the representation and depiction of humanness and the human experience, not in the exceptional, but in the mundane. By that, I mean what I believe has not changed since we became human. I explore subjects that have been examined throughout art history. The scope is vast; inspiration and ideas are not a problem. The issue is generally only a question of finding the time.

To clarify what I mean by the depictions of humans and human experience: so far, most of my installations are titled HUMAN STUDY followed by a number. The robot used in all my installations is an extremely stylized drawer; its behavior is designed to evoke human actions and, in some parts, serves as a form of self-portrait. Drawing is a uniquely human activity that we recognize immediately. When we see patterns engraved on rocks that are 70,000 years old from the Blombos Caves in South Africa, we instantly know that a human created these marks. We can almost feel the prehistoric sapiens' hand moving.

Of course, with my first installation, it is clear as I also have a human actor depicted during the performance. HUMAN STUDY #2 is a series of Vanitas, obviously focused on existence, and again includes some autobiographical elements. Here, one of three robots draws a still life featuring a skull and other objects that symbolize or evoke events and associations in the audience's imagination.

HUMAN STUDY #3 features a robot that passes the time by drawing tally marks, erasing them, and starting again, sometimes scribbling—an obviously pessimistic depiction of existence.

There are six installations in the HUMAN STUDIES series, each focused on humanness.

For the digital works, I explore the same themes in various ways and from different angles: the representation of the experience of watching or looking at something—out of a train window, at the sea, passersby, flowers, cats, human expressions, etc. Of course, there are also a good number of self-portraits.

PT: Sometimes, inspiration comes from an exhibition; an example is the series of plotted and painted robot portraits. Originally, the idea came from a show titled ROBOT PORTRAITS at the Maison d'Ailleurs science fiction museum in Yverdon-les-Bains, Switzerland.

For this series, I imagined that at some point in the future, robots would be emancipated through art and, as a consequence, develop their own culture. The subjects in these drawings would be significant in the history of robots. Some humans are also part of this fictional culture, such as Vaucansson, Ada Lovelace, Alan Turing, Ray Bradbury, and Philip K. Dick, among others. I regularly add more works to this series, including the ones you are presenting at Expanded.art.

I produced these drawings using a guided generative system seeded with images and then employed an old pen plotter that can draw with different pens. By "guided generative system seeded with images," I mean a custom software program with a graphical interface that uses an image as a seed instead of a random number. Generative algorithms are applied with different parameters on different layers, after which I can select each line or group of lines, erase them, move them, and decide on the color that will be used.

PT: Yes, always. If you think about it, a drawing is a story. For the installations, it is quite evident since they are performative and time-based, but I try to conceive them in a way that encourages different associations, memories, and stories in the audience's imagination. For example, the use of old-school desks for the robots’ bodies serves as a trigger to imagine a story around the installation. I want people to envision a narrative environment where the scene depicted in the installation is located as part of a larger story and history.

HUMAN STUDY #4: LA CLASSE, which was awarded an honorary mention at Ars Electronica, is an installation that originally featured 21 robots arranged as a classroom with a teacher and a blackboard. In this piece, the robots are strictly scripted and function as puppets. The piece consists of four acts, presenting an extremely stylized minimal story. It is about pupils learning to pass the time, comply, and conform; although they revolt, they do so in the same manner, and then they get back in line.

Jon McCormack commented on the piece in the Creative AI Podcast (S2, Ep3: Patrick Tresset, SensiLab, Monash University):

"I think it's fantastic. One thing I love about it is its humor—it's humorous yet very serious at the same time, maintaining a beautiful balance between being slightly comical and also very poignant, which is extremely hard to do. It's like watching an alien classroom or a classroom filled with machines learning, raising many interesting questions."

I didn’t mention it before, but I strive for all my works to have this atmosphere.

LA CLASSE is the fourth of the six installations in my HUMAN STUDY series. As with the other installations, drawing is an essential component. Here, mark-making is reduced to a minimal aesthetic, playing with the strong symbolic and visual contrast between the tally mark and the gestural scribble.

PT: Not at all. Nowadays, in contemporary art, it doesn’t hold much importance in my view. For me, using new technologies is not really different from when I was a painter.

PT: I don’t see this art niche as being different from contemporary art; the only specificities are the market and the way it is exhibited. Pencils, brushes, and cameras are all technologies, so all art is a combination of art and technology. Hopefully, it won’t be considered differently in the future.