Came across use of this emerging term “synthography” today in an Ars Technica piece by Benj Edwards. It seems somewhat narrowly used there in terms of synthetic AI generated photography. (Incidentally, Edwards piece also falls into what I was driving at earlier in the post about documentarians of the unreal.)

Here’s a passage quoting arist Julie Weiland:

Ars: You often call yourself a “synthographer” and your artform “synthography.” Can you explain why?

Wieland: In my current exploration of AI-based works I find “synthographer” to be the most logical term to apply to me personally. While photographers are able to capture real moments in time, synthographers are able to capture moments that never have and never will happen.

When asked, I usually refer to Stephan Ango’s words on synthography: “This new kind of camera replicates what your imagination does. It receives words and then synthesizes a picture from its experience seeing millions of other pictures. The output doesn’t have a name yet, but I’ll call it a synthograph (meaning synthetic drawing).”

Poking around, I found (in no particular order): there’s a rather complex Wikipedia page on it, that I haven’t been able to absorb yet; someone appears to be semi-squatting synthography.com (not a bad idea); Petapixel has an okay article on the topic of image-makers who work in photographic styles using generative AI tools like Midjourney – and their use of the term “synthographer” as well; this piece by Benji Friedman on synthography has some bits worth digging in on…

Friedman’s piece also appears to narrowly target photographic-style images:

The term Synthography combines the words “synthesis” and “photography.” It describes the process of synthesizing images using artificial intelligence algorithms, similar to how photographs are taken using cameras. Synthography refers specifically to AI art that involves the creation of images and animations, rather than music or other forms of art.

That seems like an artificial divide to exclude non-visual image media; I’m not sure what purpose making that distinction in a “big tent” label really would serve, but perhaps there is one I don’t see yet, being new to this term.

I don’t agree with everything in the piece, but some parts jump out at me more:

Unlike traditional art, which is often inspired by personal experiences or emotions, Synthography is driven by data.

I would disagree that this emerging genre of art, whatever we call it, is not deeply inspired by personal experience and emotion. It absolutely is driven by that, not to the exclusion, but to the cooperation of the other elements.

Unlike traditional art, which is often created with a specific goal in mind, Synthography often involves exploring the possibilities and limitations of AI algorithms and machine learning models.

I guess I see exactly that in “regular” art as well, which I’ve mentioned elsewhere in terms of interrogating technologies. This is just a description of the artistic process writ large.

Anyway, I don’t have to agree with everything in the piece for it to be a worthwhile entry in the emerging discourse around these technologies and their application to creative purposes. We need as many people approaching this from as many perspectives as possible. It’s a many worlds situation to begin with. Welcome to the Multiverse.

Related writing & research