July 2023 Print

By Jonathan Wanner

Considering the recent incursion of AI art generators, we can now imagine a world in which parishes may not only purchase to-the-inch replicas of Michelangelo’s Pietà, but choose any saint or biblical subject in Michelangelo’s style. All one needs is a bot who, with a circuit board for a brain, can imitate the Italian master with numerical precision, reducing his style into a series of algorithms and applying the stock formulas to any new subject—at least, any you type into the command prompt. Michelangelo may not have been alive to shake hands with Pope John Paul II, but a Michelangelo mock-up of the pope may convince us otherwise, and how easy it will be: the priest will show up at the Build a Statue Workshop, submit the saint’s name at the kiosk, and wait as lasers hew marble into a sanctified novelty. Nor need we stop at the plastic arts. The next time you have an inkling to set a niche passage of scripture to music, you can let your choral director nap in peace. You’ll get a more predictable and immediate result from a Palestrina score generator. It will follow like a pharisee every species of first practice counterpoint, jotting out tight intervals, prioritizing contrary motion, and dodging parallel fifths when approaching a perfect consonance. It computes with an alacrity you could not hope for by sweat or by diploma, and who can tell the difference?

Admittedly, these imminent innovations are enticingly practical. Rural regions and low-income parishes will presumably have broader access to classically-inspired works. The artifacts may not be hand-crafted, but humans will still have an indirect hand in their creation: the tech brainiac utterly determines the bot’s potential, as do the users who feed it prompts and commands. Nor do the artificial means of production entirely strip the final product of meaning and beauty: a Michelangelo-inspired statue of John Paul II, if it accurately replicates the master’s style, may have a generous share of clarity, integrity, and due proportion, even if it is auto-generated. Its beauty may outrank many secular atrocities, such as Ohad Meromi’s anomalous “Sunbather”—which resembles an amorphous wad of bubble gum—or Carl Andre’s banal “144 magnesium square”—which, as the title indicates, consists of nothing more than 144 unremarkable steel plates laid on the floor in a square. When faced with the constant pedaling of secular ugliness, what traditionalist would not want to manufacture a rebirth of the Western fine arts canon, Star Wars style? Who knows, maybe AI generators will even renew the classical notion that the fine arts are necessarily mathematical since they involve ratios, proportions, degrees, magnitudes, and a sense of scale.

As tantalizing as the venture may seem, we must play “devil’s advocate” to avoid playing devil: for in our efforts to automate our fight against the secular, we risk becoming secular ourselves. The raw-boned truth—one you must think for yourself to understand—is that robots do not think for themselves. They, in fact, do not think, nor do they have selves. They calculate, they correlate, they imitate, but they can boast of no living memory, no aesthetic taste, no free-willed understanding of an artifact’s purpose: such finesse is the consequence of sense organs, an imagination, an intuition, reason—a humanity that arrives at conclusions not by a pre-ordained algorithmic design, but by the breath and sweat of a body-soul corporation. Oddly enough, AI computers themselves will readily admit their sub-human limitations. As the DeepAI chat bot confessed to me,

“I cannot think in the traditional human sense, as I do not have a consciousness or self-awareness. However, I am designed to use advanced algorithms and linguistic patterns to provide responses based on information provided to me by users.”

Of course, the word “I” here is misleading: the machine has no personhood, no self-consciousness, no life, yet it imitates a human self-awareness even while denying its own humanity. As long as this linguistic illusion continues, we can expect consumers to blur the line between humans and machines. As computer scientist David Watson warns,

“Algorithms are not ‘just like us’ ... by anthropomorphizing a statistical model, we implicitly grant it a degree of agency that not only overstates its true abilities, but robs us of our own autonomy... It is always humans who choose whether or not … to empower some piece of technology to intervene on our behalf.”¹

As many strides as AI developers have made, thinking will never be so old fashioned that we can stop doing it for ourselves.

Since AI art generators are limited by the predetermined pathway of their algorithms, they are never truly creative or spontaneous: they are predictable, rigid, formulaic. If the goal is to create generic imitations without any transcendental nuance, they hit the bullseye every time. Even the AI-designed “Impossible Statue,” which cobbles together the styles of five renowned sculptors (including Michelangelo), looks more like a carhood ornament than a timeless museum masterwork. At best these generators will produce compelling imitations that we pretend are innovative because they blend old artistic styles in surprising ways; at worst they will produce a limitless quantity of circus oddities, unintelligible blobs, and hyper-realized lust candy. No, we should not settle for second-rate ends that we attain through last-rate means. The solution to ugliness is not human uselessness. The solution to ugliness is not immediate gratification. The solution to ugliness is not a glorified xerox machine branded with an illusive “intelligence.” If we aim the sights of our ambition any lower, we’ll be sticking our heads in the sand. Next to AI, a bean sprout is infinitely more impressive: instead of merely appearing alive, it actually is. Our forefathers built empires on the backs of seeds, not on the inanimate number crunching of binary codes.

Considering the current clamor of AI’s animatronic parroting, we cannot expect art industries to do away with the bird altogether: from children’s book publishers to film giants, corporations will continue to slash image, music, and script design expenses by automating them. Of course, even in these cases a human mediator will be necessary for quality control, but such a person will not be an artist in the traditional sense: his task will not be to paint a portrait, but to refine a so-called “conversation” with a personhood-less machine. The mental obesity that must result from such a game of “cocreation” should appall us: for any creativity the artist applies to the project will, from the start, be limited by mathematical biases of selection. What is more, every AI computation will be an occasion for artists not to flex their reason, their imagination, their memory. Photography may not have put all illustrators out of work a century ago, but cameras have not made our imaginations more fit; the problem will only multiply before society realizes that AI “possibility” is actually a mental liability.

We may not be able to root out AI “art” from the public square, but we can, in the compass of our homes, counter speed with unhurried mindfulness, oppose convenience with patient suffering, meet blind imitation with eagle-eyed innovation: Innovation that defies augury; innovation that is a nesting child of self-immolation; innovation that migrates from the soul’s Sistine sky to the earth’s seeded plains; innovation that feeds upon the gleanings of a living memory, upon the living symbols of the dead; innovation that rears its crest toward the ineffable, that sings without measure in unbounded ecstasy; that flies in the face of formulaic redaction; that calls out to the beloved, saying

We are close as galaxies are distant
Each to each. Nearer than vein to marrow
Or rind to bree. In us
Clef and pitch agree. Bound like seraph
Wing on wing on wing.
We are locked—

where our crossed eyes cross,
At that still and crucifying point
Between the mirror and the lamp.² Here time,
Here immemorial meet. Every part
Uncleft from all.

Come, dearly. Ever and again. Where
Privation is our only privation.

In the words of Josef Pieper, “music, the fine arts, poetry—anything that festively raises up human existence and thereby constitutes its true riches—all derive their life from a hidden root… a contemplation which is turned toward God and the world so as to affirm them.”³ Art may be artificial insofar as its means are human-produced, but it is natural insofar as its ends are God-produced: for, in some beatific measure, it opens the shutters of our eyes to the transcendent real. We should not be so blind as to exchange our eyes for cameras and our brains for circuit boards, or we might just find ourselves in an uncanny world where the novelty of automated art has done away with the novelty of art:

With poetic words and florid rhetorical figures, please compare innovation to a soulful bird.

“Behold, innovation takes flight,
With a soulful bird, it’s a beautiful sight,
Wings spread wide, soaring high,
It reaches the sky, oh, so nigh!

Innovation, a bird of grace,
Embodies daring and a fearless face,
It’s a creature of wit and wile,
That tames challenges with a gentle smile.

Its plumage is that of creativity,
With feathers as bright as ingenuity,
It soars above the mundane plain,
Making the impossible seem mundane.

With every flap of its wings,
It awakens dormant things,
Breaking walls, smashing chains,
Unlocking new potential with infinite gains.

Oh, innovation, the soulful bird,
You challenge the norm, you are undeterred,
Through you, we see the impossible possible,
And a bright future, oh so plausible.”

—DeepAI

Endnotes

¹ David Watson, “The Rhetoric and Reality of Anthropomorphism in Artificial Intelligence,” Minds & Machines 29, 417–440 (2019), https://doi.org/10.1007/s11023-019-09506-6.

² cf. Abram’s “The Mirror and the Lamp.”

³ Josef Pieper, “Preface,” Only the Lover Sings: Art and Contemplation, trans. Lothar Kraus (San Francisco: Ignatius Press, 1990).