The Nu-Normal #20: A Portrait of the AI as an Artist
The Nu-Normal #20: A Portrait of the AI as an Artist
A consideration on the blurred ethical lines behind AI artwork.
Back in August, I wrote about the phenomenon of retroactive editing and its impact on considerations of art:
My conclusions in that piece largely hinged on weighing up two different philosophical positions, namely those of artistic integrity and aesthetic sincerity. Ultimately, I felt that while we might quibble about whether creators should do retroactive editing or not, the rationale behind committing the act itself is not insincere at the artistic level.
Or, as C. Thi Nguyen so eloquently put it:
“In practices where we pursue originality, there can be no guarantee that an artist will find an audience or that audiences will find art to their liking. But that is the cost of engaging in any deeply unpredictable creative endeavor. The best we can do is to collectively adopt commitments to aesthetic sincerity—to promise each other to be guided by our sense of aesthetic value. We cannot coordinate on a specific aesthetic result if our interest is in finding new results, but we can coordinate on staying together within the realm of the aesthetic.” (48)
However, there’s one issue that appears to be complicating this notion of aesthetic sincerity, particularly when it comes to the production of artwork for commercial purposes. So, let’s discuss the blurred ethical lines of treating AI as an artist.
1. “It is the reflection of a basic reality”
Earlier this week, Pennsylvanian metalcore veterans August Burns Red unveiled their ninth studio album, Death Below, set for release on March 24th through SharpTone Records:
According to lead guitarist JB Brubaker, Death Below is “the darkest, most personal album,” inspired by the “extremely dark and challenging” events of the years following the release of their previous full-length, 2020’s Guardians. The album’s press material states that the record “encompasses all of the pissed-off strength of the band’s past without sacrificing an ounce of catchy, melodic might.”
All is well and good so far, then. I certainly enjoy August Burns Red, and their new album’s lead single, “Ancestry,” is suitably fun and heavy. And while I haven’t really latched on to much of their musical output from the 2010s, I still regularly bump tracks from 2007’s Messengers in my various gym playlists. (“Composure” still absolutely goes, by the way, in case you’re wondering.)
However, what stood out most from the discourse surrounding the album announcement was the band’s choice of artwork: a piece titled “The Valley of not fully formed seating Demons” by UK conceptual artist Andrei Riabovitchev, who posted an earlier version (along with the additional vinyl variants) back in May:
2. “It masks and perverts a basic reality.”
Now, I can already hear what you’re saying. “Okay, what’s the big deal? It’s grim, evil-looking artwork for a metal band. Hardly newsworthy.”
The issue, I think, stems from the fact that Riabovitchev’s original post includes the following tags: “#aiartwork #AIart #midjourney” Here, we have a professional artist with a strong portfolio of original* work listing art created using Midjourney, which is then—I would assume—licensed to August Burns Red at a later date for commercial use in the release of Death Below.
Thankfully, I wasn’t the only one to notice this fact. On Reddit, fan opinion swung from “I dig it” and it’s “abstract” to “this is not one of their better album covers imo. super messy looking.” The denizens of Twitter, on the other hand, were not as receptive:
For the purposes of our discussion today, then, we should focus on this: Who, in the case of “The Valley of not fully formed seating Demons,” is the real artist?
3. “It masks the absence of a basic reality.”
One of the people chronicling the ceaseless rise of our eventual robot overlords has been journalist Ryan Broderick. For much of this year, Broderick has been highlighting the increasing intensity with which AI-enabled applications encroach on creative spaces over at his excellent internet culture newsletter, Garbage Day.
In terms of AI-generated art, Broderick has reported on the creation of Discord servers to share prompts for “generative” art, in effect becoming de facto meta-marketplaces for the tools necessary to create intricate AI artwork. Elsewhere, this type of activity is leeching into creepier spaces, with weird and horny NSFW fetishists spending hours in Midjourney perfecting just the right prompt required to simulate the perfect curve of an anime girl's buttock.
All of this is to say that the creative line between artist and tool is becoming more blurred by the day. And for Broderick, there’s a tipping point on the horizon that’s almost inevitable:
“Eventually, an A.I. is going to make something that humans become obsessed with that only an A.I. can make, and then we will find each other in very strange new territory.”
To help answer our question, Broderick points to a fascinating tweet thread from illustrator Douglas Bonneville, who documents the process he used with the Stable Diffusion AI to generate a base artwork, which was adjusted and edited into something resembling a “traditional artwork”. However, what’s really of interest, I think, is the poll Bonneville throws up towards the thread’s end:
For me, these four questions get right to the burning heart of the issue. And from 357 votes, over 90% of people appear to agree that Bonneville’s AI-assisted piece is, in fact, art (comedic value, notwithstanding).
Now, before we get carried away here, does this result have more to do with Bonneville’s already cultivated audience and his established position as a professional artist and illustrator with a marketable skill set outside of using AI prompts? Perhaps. And I do think it’s worth considering how these results may differ if this thread was published online by regular Joe Citizen.
4. “It bears no relation to any reality whatever: it is its own pure simulacrum.”
One of the most thoughtful approaches to our question comes from the Cambridge philosopher of consciousness, Henry Shevlin. In a piece titled “The artist is dead, AI killed them: Can AI be genuinely creative?” Shevlin considers what impact the human has on the creation of art when a new technology can replace certain skills:
“A lingering question may persist in the minds of readers, however, as to whether there is something fundamentally limited about the capabilities of these systems. A gifted human artist like Monet or Picasso can move beyond the constraints of the art world they inhabit and create a bridge to truly novel forms of representation. Could an image model ever do that, or does their widespread adoption instead presage an era of artistic stasis and stagnation, in which the heterogeneity of the human art world is flattened, normalised, and ossified by the statistics of its dull machine-creators?”
To arrive at a conclusion, Shevlin leans on the work of philosopher and cognitive scientist Margaret Boden, who breaks down the capacity for creativity into three distinct categories:
“The first is combinatorial creativity, the rearranging of existing elements to create something new; the second is exploratory creativity, finding novel ideas or forms within existing paradigms; and the third is transformational creativity, which involves the creation not merely of a new work or idea, but of a wholly new artistic framework or way of approaching a problem.”
Shevlin makes a convincing case for AI-generated art exhibiting the first two categories but holds back on the third. So, let’s apply our August Burns Red case study and see where we land:
Does “The Valley of not fully formed seating Demons” rearrange existing elements to create something new? Arguably yes. However, without knowing the exact prompts used, this is hard to say definitively. I think we can assume, based on how these applications work, that Riabovitchev likely used existing prompts (i.e. words) and generated new work from them.
Does “The Valley of not fully formed seating Demons” find novel ideas or forms within existing paradigms? Again, yes. Using combinatorial creativity, the artwork expands on images and concepts like “valley” and “demon” and transforms them into grotesque proportions of size and intensity.
Does “The Valley of not fully formed seating Demons” involve the creation not merely of a new work or idea, but of a wholly new artistic framework or way of approaching a problem? Sadly no.
Here, I agree with Shevlin in suggesting that the jury is still out on the true creative limitations of AI art generation. It’s clear that Riabovitchev is an artist, but it’s not entirely clear if “The Valley of not fully formed seating Demons” should be classified as his artwork without knowing the full amount of editing and transformation undertaken once the images were initially generated (i.e. as documented in Bonneville’s thread above).
As for August Burns Red and the discourse surrounding their decision to use “The Valley of not fully formed seating Demons” as their album artwork, this largely reduces to means and motive, considerations that cannot be answered by purely perceiving the image in of itself.
Could they have afforded to pay a human to create something similar? Undoubtedly. Are they trying to make a statement by choosing not to do so? Debatable. Ultimately, I think the band thought the artwork looked cool, and then they paid to use it. Case closed. And on that note, I’ll leave you with some wise words from Bonneville:
