In museums, I will sometimes take macro shots up close of famous (and not-so-famous) paintings. Partly that’s interesting because you can see some of the material details of the techniques used to make the painting—the brushstrokes, the pigments, etc. are much more visible. Partly I’ve found that you can sometimes isolate a part of the painting that effectively creates a new derivative artwork. I once did a series of medieval painting macros that I thought of as a possible set of avatars for asynchronous threaded forums.
This photo looks pretty cool but it’s just an unthoughtful reframing of an artwork at the Los Angeles County Museum of Art. No creativity on my part at all. I’m not even sure that my tight framing here is a good idea or that it works especially well, other than disguising the full look of the installation at LACMA.
I have been thinking more this week about the visual side of the current generation of AI—I’m hoping to play around with Midjourney some more soon. I don’t think it’s the end of visual creativity or originality, but it’s clear that anyone looking to create artwork of any kind, including photography, is going to have to think in new ways about human style and expressiveness. And that we’re going to need new ways—maybe blockchain-enabled—to register our creations, because the first moment that someone makes something that is not just “in the style of an existing artist” is going to be about five minutes before someone trains the descendants of current AI on that new work so that it can be added to their repertoire of styles.
That’s new in scale, it’s new in speed, and it’s new in terms of the specific outcomes, but we need to remember that much of the older visual art we admire was made in studios that churned out incredibly similar work as a service to the nobility, the gentry and the clergy. The proposition that we have lived under for the last two centuries or so, that art should be made by individuals whose distinctive creativity is made manifest within it and valued as a commodity to the extent that it does so, is not a universal proposition about art at all.
We may find with visual culture that having AIs in a kind of atelier with a verbally or textually skilled overseer who prompts them this way and that opens up something surprising about the visual imagination of many people, much as the Internet revealed a kind of rich deep layer of creativity that older media with their gatekeepers failed to tap into or demonstrate.
But anybody who has done the kind of brute-force derivation of visuality from other people’s visuals of the kind on display in this photo is, I think, out of luck if that’s been a paying gig for them up until now.
An artist friend of mine just found that some of her work has been “scraped” for use by AI. She is not thrilled to be an unwilling part of the AI “atelier,” Tim. Artists can’t escape putting their work up on various websites if they want to publicize their art these days, but it is these same sites that are being used as fodder for AI “creativity.” So what about the small scale artists who provide the fodder for these second-order users of AI? Should they just resign themselves to feeding the machine, or should they commit commercial suicide and take their work completely off-line? There’s a technophilia in your recent posts that I’m wondering about. I’m not ready to go post-human quite yet.