Update, 7/15: Bumped up as long-ish weekend read and conversation for the weekend!
Jeffrey Berg, the artist and technologist who became famous as the Metaverse artist known as "AM Radio", a Second Life avatar who created profoundly surreal and poignant immersive installations like "The Faraway" (above), a rusty train lost in the wheatfields, recently discovered a surprising thing:
It is now possible, using programs like Craiyon, a text-to-image generation tool powered by AI (a variation of OpenAI's DALL-E), to automatically create images strikingly similar to the Second Life installations which took him many hours if not days to create in 3D.
See below:
"Was messing with Craiyon today and I wondered the impacts it could have on metaverse space creation," as he mused to me recently, sharing the results above. "3D geometry rather than a generated image from plain text input doesn’t seem like a huge stretch."
That started a long, fascinating, and -- this being AM Radio, after all -- moving conversation with Jeff on the future of art in the era of DALL-E, AI and machine learning in general, and promising AI-based art like that by Dr. Nettrice Gaskin:
Above: RL acrylic painting by Jeffrey Berg
WJA: What do you think about the implications [of AI] as an artist?
Jeffrey Berg: I’ve been thinking about it a lot… If we were in an elevator this is what I’d say in 90 seconds: The current moment of machine learning in 2022 is the novelty itself, not unlike the introduction of glow, or WindLight, even mesh in Second Life, where so many creators in SL could not overcome being merely presenters of the technological capability rather than its master. We’re seeing that with machine learning, where views and viral success is often linked with a mere gaze into the uncanny valley rather than anything a human is trying to express.
I’m hesitant to jump to the use of artificial intelligence just yet, which is a whole topic on its own. I find most of the debate cliché and lost in what i would call a pop perception of ML more closely related to pop-psychology than the implications of ML statistical algorithms at scale.
At the end of the day, no matter what the machine produces, what we ask it to or some genetic algorithm is used to formulate an output from a machine, it is still for humans to judge the artistic achievement. And by artistic achievement I mean declaring a new rung towards the endless pursuit of the sublime and search for those boundaries of existence as humans we no longer wish to tolerate. That is art.
For the creator then, do we present ML, or do we wield it? Can I explore even the Duplo blocks of things like Tensorflow and toys like Craiyon and create a strange experience for people that makes me wonder about the technology, or can I create an experience that reminds me how human we are and possibly expand the boundaries of the definition of what it means to be us.
I have always believed at a core level that first emotive communication between humans is a cry for empathy, the moment we are born we scream help me, I’m alive. What an amazing, beautiful, and terrible moment for us all.
This cry for empathy continues, as it always has since the first cave painting. This is true especially as artists, as we gaze into the world and simply ask if anyone else is seeing what we’re seeing too, and importantly we ask “do you feel it?”
As we try to even define the it, the indescribable as if we’d discovered and trying to explain a new color, a new way of being. That gaze includes that scary and gripping view into the uncanny valley of machine learning. People talk about the death of creativity and art in light of ML, and yet I sit in awe at the work ahead for every artist to explore and uncover the virtual as it begins to gaze back at us.
WJA: Maybe it's just me but when an image is described as AI generated I immediately shut down emotionally/intellectually, or I think of it in another category that's not art.
JB: Yeah, you’re in the thick of it with your subject matter expertise, the novelty wore off quick, like a Renaissance mechanical man or other machinery that could mimic human actions seems enigmatic but limited to us now, but likely captured the audience of the time.
A cave dwelling human from long ago could gather up some fellow dwellers and say look when I spit out this charcoal, I can recreate my hand on the wall! It must have been astonishing and frightening but eventually boring.
What makes those cave paintings so enigmatic and long lasting is that it wasn’t the invention of painting on a surface, it was the day someone said I can use this to augment language and tell a story, a story that will remain even after we are gone. Suddenly the communication becomes forefront, not the technology, and the medium becomes the ability to express information and emotion across space and time.
Artists will use machine learning to tell a story, a much bigger one than where a good place to hunt is, but the one that reminds us how human we’ve always been. I would look for those people, the ones saying look what I found in the sea of virtual serendipity called ML, and declare it to be astonishing and dreadful and and maybe even a little sad but there in the soup there is hope and inspiration too, as all successful art has always done.
Machine learning is a part of our world if we like it or not. We can compare an output of Craiyon to an image of an atomic bomb and say look, this will destroy us. Or we can see it as an expansion of our existence, this thing we’ve unlocked that beckons us to explore it and tell its story, and call that story art.
And in regards to comparing ML to the atomic bomb, we can look at game theory and the discourse at its core to find lessons.
To quote John Von Neumman, the mathematician and game theory polymath (via Healey Donovan’s Fortune magazine notes of 1954) “The world in general is, and the U.S. in particular, is riding a very fine tiger. Magnificent beast, superb claws, etc. But do we know how to dismount? […] you are pessimistic because you cannot yourself visualize any form of world organization capable of controlling the world’s instability and ending the danger. And you haven’t heard of anybody else who’s visualized that organization.”
And yet it was never the technology, never the bombs, just waiting underground, sitting on top of thousands of pounds of explosive fuel to be delivered through space to provide the worst spectacle ever. We accepted that possibility. The story however was the human one, how we dealt with our new circumstance, and how we continue to do so today. There are countless art achievements, even sublime ones that tackle the atomic age.
Machine learning fills us with fear, it makes us turn away, we dismiss it. That is exactly the marker for an artist to plant a flag and say it’s time to turn towards this and find out how very superb but terrifying this fine tiger really is.
Kendrick Lamar Deep Dream neural network portrait copyright Nettrice Gaskins.
WJA: You've seen Nettrice's Deep Dream portraits? I think that's a good application of art + AI. [Read more about her work here]
JB: Absolutely, to me, it’s most important implication is not so much the end product, although beautiful, the work puts the learning bias of race to the forefront, compounded by systemic racism’s affect on the digital divide and unequal access to advance tech tooling such as ML.
In the work, suddenly it’s no longer about ML, but about the story of our culture and society in its interactions with technology.
Realistically, systemic racism means that most people of color will encounter machine learning in the face of racial profiling by machine learning algorithms. If it’s during a protest or a simple traffic stop, this is the reality for many, ML and data in general is a source of fear. This inexcusable disparity of experience is highlighted by this far different experience of Afrofuturist possibilities that exist despite the bias. On this case It’s not a story about the novelty of the technology, it’s about human beings and the dilemmas we face and must face up to as ML models reflect even unconscious bias in very forward and blunt ways.
The art won’t be easy, but to quote Interstellar, it’s necessary.
More reading on AI and art by Jeff Berg here:
Very insightful post. Craiyon is not really doing justice to the real thing. AI generated images are just starting and have a long road ahead of them.
Posted by: Count Burks | Tuesday, July 19, 2022 at 10:45 AM