Few technologies have demonstrated as much potential to shape our future as artificial intelligence. Specialists in fields ranging from medicine to microfinance to the military are evaluating artificial intelligence tools and exploring how they could transform their work and their worlds. For creative professionals, ai poses a unique set of challenges and opportunities, particularly generative ai, the use of algorithms to transform large amounts of data into new content.
The future of generative ai and its impact on art and design was the topic of a sold-out panel discussion on October 26 at the MIT Bartos Theatre. It was part of the annual meeting of the MIT Arts Council (CAMIT), a group of alumni and other supporters of the arts at MIT, and was co-presented by MIT Center for Art, Science and technology (CAST), an interscholastic initiative for artist residencies and interdisciplinary projects.
Presented by Andrea Volpe, director of CAMIT, and moderated by Onur Yüce Gün SM '06, PhD'16, the panel featured multimedia artist and social science researcher Ziv Epstein SM'19, PhD'23, professor of architecture from MIT and director of the SMArchS and SMArchS AD programs Ana Miljački and artist and roboticist Alex Reben MAS '10.
The discussion focused on three themes: emergence, incarnation and expectations:
Appearance
Moderator Onur Yüce Gün: In much of your work, what emerges is often a question, an ambiguity, and that ambiguity is inherent to the creative process in art and design. Does generative ai help you achieve those ambiguities?
Ana Miljacki: In the summer of 2022, the Mostar Memorial Cemetery (in Bosnia and Herzegovina) was destroyed. It was a post-World War II Yugoslav monument and we wanted to find a way to defend the values that the monument had stood for. We collected video footage from six different monuments and, using ai, created a non-linear documentary, a triptych that plays on three video screens, accompanied by a soundscape. ai-video-installation/”>With this project We create a synthetic memory, a way to plant those memories and values in the minds of people who never lived those memories or values. This is the kind of ambiguity that would be problematic in science and is fascinating to artists, designers and architects. It's also a little scary.
Ziv Epstein: There is some debate about whether generative ai is a tool or an agent. But even if we call it a tool, we must remember that tools are not neutral. Think about photography. When photography emerged, many painters worried that it would mean the end of art. But it turned out that photography freed painters to do other things. Generative ai is, of course, a different type of tool because it relies on a lot of other people's work. There is already an artistic and creative agency built into these systems. There are already ambiguities about how these existing works will be performed and what cycles and ambiguities we will perpetuate.
Alex Reben: I am often asked if these systems are really creative, in the same way that we are. In my own experience, I am often surprised by the results I create using ai. I see that I can direct things in a direction parallel to what I could have done on my own, but that is different enough from what I could have done, amplified, altered or changed. So there are ambiguities. But we must remember that the term ai is also ambiguous. In reality they are many different things.
Incarnation
Moderator: Most of us use computers daily, but we experience the world through our senses, through our bodies. Art and design create tangible experiences. We hear them, we see them, we touch them. Have we achieved the same sensory interaction with ai systems?
Milacki: While we work on images, we will work in two dimensions. But for me, at least in the project we did around the Mostar memorial, we were able to produce affect on a variety of levels, levels that together produce something that is larger than a two-dimensional image moving in time. Through images and a soundscape we create a spatial experience in time, a rich sensory experience that goes beyond the two dimensions of the screen.
Reben: I guess for me embodiment means being able to interact with the world and modify it. In one of my projects, we used ai to generate a “Dali-like” image and then turned it into a three-dimensional object, first with 3D printing and then casting it in bronze in a foundry. There was even a patina artist to finish the surface. I cite this example to show how many humans were involved in creating this work of art at the end of the day. There were human fingerprints at every step.
Epstein: The question is how do we build meaningful human control into these systems, so they can become more like, say, a violin. A violinist has all kinds of causal inputs: physical gestures that he can use to transform his artistic intention into results, into notes and sounds. Right now we are far from that with generative ai. Our interaction basically consists of writing a bit of text and receiving something in return. We are basically yelling at a black box.
Expectations
Moderator: These new technologies are spreading so quickly, almost like an explosion. And there are enormous expectations about what they are going to do. Instead of hitting the accelerator here, I'd like to test the brakes and ask what these technologies aren't going to do. Are there promises that you won't be able to keep?
Milacki: I hope we don't go to “Westworld.” I understand that we need ai to solve complex computational problems. But I hope it is not used to replace thinking. Because, as a tool, ai is really nostalgic. It can only work with what already exists and then produce probable results. And that means that it reproduces all the biases and gaps in the archive that it has fed. In architecture, for example, that archive is made up of works by white male European architects. We have to find a way not to perpetuate that kind of prejudice, but to challenge it.
Epstein: In some ways, using ai now is like putting on a jetpack and a blindfold. You're going very fast, but you don't really know where you're going. Now that this technology seems to be able to do things similar to humans, I think it's a great opportunity for us to think about what it means to be human. My hope is that generative ai can be a kind of ontological wrecking ball, that it can change things in a very interesting way.
Reben: I know from history that it is quite difficult to predict the future of technology. Therefore, trying to predict the negative (what might not happen) with this new technology is also almost impossible. If we look back at what we thought we would have now, at the predictions that were made, it is quite different from what we actually have. I don't think anyone today can say with certainty what ai won't be able to do one day. Just as we cannot say what science or humans will be able to do. The best we can do, for now, is try to push these technologies into the future in a way that is beneficial.
