Tuesday, March 05, 2024

Is Artificial Intelligence (AI) Bad for Sacred Art?

I have been asked this question many times recently. Most who ask it are worried that it will have a detrimental effect on the sacred art that we see in our churches and homes. My response is that while AI may pose many great problems in other aspects of human life, I think that its impact on sacred art will be either negligible or positive.

An AI generated Simeon and Christ
I don’t believe that it is possible for AI to create art that is of the right quality, no matter how good the technology is, because authentic sacred art relies on inspiration from God for its originality. AI is not open to the promptings of grace as human beings who possess an immortal soul are. (While God is omnipotent and might, in principle, miraculously influence an AI process, I think this as unlikely!) One of the reasons for the decline in art in the last 200 years, in my opinion, is that people have forgotten the importance of the supernatural in creating beauty. AI relies on replacing natural human faculties, albeit in a powerful way. This is highly speculative, but this might mean that it could be good at generating better art in the inferior styles that emerged during this latter period. 

Every tradition must reinvent itself anew with each generation in order to be vital and to sustain itself. The assumption in the past was always that unless the artist is open to God’s grace, he will be imitating only what artists have done in the past, to some degree or other, and the quality of art will decline. So even beautiful work made in precise imitation of the style of the past will not resonate with large audiences today.  

The test of my hypothesis will be in the quality of the art produced. Here are some examples I found online. There is one that was described as ‘Byzantine style’, clearly inferior to the work of contemporary iconographers.

This one is in a naturalistic style which looks to me like modern versions of 19th century realism.  

They are comparable perhaps with the sacred art that I see from people trained in the academic method who don’t understand the principles of sacred art. Their work always looks sentimental and superficial, despite the skill in execution, just like 19th century realism does to me. I wouldn’t choose it, but many Catholic churches around the US are commissioning art in this style at the moment, to my disappointment. Perhaps some might decide to go AI instead, assuming that it is possible to create a finished piece that is actually a physical painting, and not just an electronic image.

Art is as good as it looks, and I will be proven wrong if we see computer generated art that is spectacularly good by any measure, and is indistinguishable from the very best art made by man under inspiration from God.

Let us imagine hypothetically, that this happens. Wouldn’t that be a good thing? It means we have lots of original and great art drawing people to the contemplation of God, all produced at low cost. Admittedly, it might put some artists out of work and consign the mediocre artists, like me, to the dustbin, but that’s not the end of the world.

The point of having art is not to employ artists, but to draw us to the contemplation of God. As with the argument with Luddite weavers and spinners in the 18th century, a new spinning machine might put hand spinners out of business, but the result is better yarn for all at cheaper prices, and an overall benefit to the well-being of society as a whole.

One common fear that I hear is that it is important to preserve the relationship between the artist and the viewer that is created through the mediation of the work of art. By looking at the art, we get insights into the nature of the artist, and this is an important part, it is argued, of the whole reason that we have art.

I don’t accept that this is a worry, because I don’t agree with the premise. It is the Romantic mindset, and not the traditional Christian one, that is concerned about the supposed relationship between the artist and the viewer. The goal of the good artist is to be invisible to the viewer. I would argue that all art has a clear purpose, which is to imitate nature, and through that, to draw people to the contemplation of the image and ultimately through the image, to contemplate God, in a way that is in accord with God’s governance. The only way that this would properly involve consideration of the artist is if the painting is a self-portrait! Even a simple landscape can lead us to the contemplation of God by highlighting the beauty of creation.

Once a piece of art has been completed, questions about how it can be made immediately become irrelevant. In the traditional Christian way of looking at things, we are only interested in how well it fulfills its function.

In short, art is as good as it looks. If there is more and better art by this measure as a result of AI, then that is a good thing, and I would be happy to see it. It seems similar to me to the situation that arose when photography was invented, or when musicians could start to use computers to create music. Both are relatively new media, and can be regarded as tools which can be used well or badly, and haven’t yet replaced painting or musical instruments. Rather, they have enriched art and music by increasing the range of what can be produced in the hands of a skilled photographer or composer.

Perhaps in this new situation, we will give the person who asks Chat GBT the right questions, or even the programmer who created it the right questions the credit for being the artist!

Below, painted and photographed portraits of American Presidents, Jefferson and Lincoln.

More recent articles:

For more articles, see the NLM archives: