At first blush, it may seem that sacred art and artificial intelligence have nothing to do with each other – after all, they seem to inhabit two entirely different spheres. However, as tools and methods change and evolve, even very traditional work is being improved by technology, whether it be farming or sewing or woodwork or what-have-you.
The St. Elisabeth Convent in Minsk is an Orthodox Christian convent that has a large internet presence. Not only do they have their website – which has versions in English, Russian, German, French, Italian, and Serbian (but not Belarusian, unfortunately) – but they have Instagram, Facebook, various YouTube channels, Twitter, Telegram, and LinkedIn . (I can’t vouch for the other languages, but the English version is surprisingly well-done. ) In any case, I guess I shouldn’t be too surprised, but it seems as if they have a Reddit account as well.
About a month ago, there was a post made by this account with the question “What do you think about icons made digitally?” The post featured the art of a young woman who creates art in Photoshop who donated a couple of works to the St. Elisabeth Convent. While neither of the works attached to the post would ever be considered icons, a later post by the St. Elisabeth Convent, there is an example of a “digitally painted” icon.
The thing is, I don’t think it’s possible to say that computerization must stay out of this type of art. I think it’s already here, in many ways, from computerized cutting, to creating grids and templates, to mass-producing the images that get glued to boards to create non hand-painted icons. If every church could only have hand-painted icons, there would be so many fewer because the price is so high considering the materials and time required for each one.
I believe that a major argument is that computerizing the creation of icons would “cheapen” the process; the idea being that people can just throw something together in Photoshop and call it good. However, mastering Photoshop is something that few people actually do. I would guess that most people who use Photoshop use it to alter images rather than create them. After watching this video of the artist of the digital images, it’s impossible to say that there isn’t a lot of work and artistic talent going into the work:
That being said, at the end of the day, a piece of art like this doesn’t exist in the physical realm the way that an icon hanging on a wall does. What has been created is a rearranging of bits and bytes on a tiny piece of computer storage media. Again, fine and good, but does that also mean that a book isn’t a book until it has been physically printed or photos aren’t photos unless they have been printed or exist on film?
Extrapolating even further, then, what about AI Sacred art? Or can such a thing even exist. Does sacred art require some semblance of human authorship as is required for the US copyright office? Is the human authorship crucial because of the attitude we have vis-a-vis our relationship to and with God? In a negative sense, that would say that because a computer cannot be in relationship with God, it cannot perceive any sense of the sacred, and so as good as an image might be, it cannot be “sacred”.
As a practical matter, too, more and more tasks are being handed over to AI to “solve”. There are significant questions, though, as to how much of the answers are actually due to real artificial intelligence, and how much is due to how much information can be searched through in an astonishingly small amount of time. Two examples of this – there seem to be a proliferation of tech “review” sites that are just thrown together by AI programs and at best quickly looked over. At first glance, they seem to be okay, but the language 1) seems somewhat superficial and as though not written by a native speaker and 2) different sites seem to have almost the identical odd writing. For example, in looking for camera lenses, I thought it was very weird that a bunch of “review” sites seem to have primers for taking pictures and sentences that sound like they were cribbed from sales pitches. A normal person giving recommendations just doesn’t talk or write like that. My second example is with voice recognition over the phone – I have had a lot of trouble with systems that can’t filter out background noise. All I need is one child to start calling “Mommy, mommy!” loud enough for the system to pick it up to throw my call out of whack. Never mind if a kid starts crying in the background – a human on the other end would tune it out, but I’ve actually had automated systems hang up on me because they couldn’t filter out the noise in the rest of the house. It just seems that if “artificial intelligence” has so much trouble with “easy” tasks such as these, the propensity to really botch something that is supposed to hold more than an artistic meaning is too much of a risk to allow it to run freely.
I think the key here lies somewhere in understanding that even in sacred art, technology of different types is a tool, but the tool cannot be left to run on its own. In some ways, it reminds me of the short “The Sorcerer’s Apprentice” from Disney’s Fantasia . Mickey is tasked with bringing water at the behest of the sorcerer, but when the sorcerer leaves, Mickey enchants the broom to do this work for him. He falls asleep now that he doesn’t have to do the work, and when he wakes up, there’s trouble, and his panicked solutions only make the situation worse. If we use technology as a tool, something created to serve its creator(s) (or THE Creator) it isn’t necessarily a bad thing. However, there’s a line between using tools to do things better and , in our laziness, expecting these tools to make decisions that require “human authorship”