Skip to main content

Recently, there’s been even more press than usual about AI proliferation and its associated risks. The hype has been driven, in part, by the now infamous Ross Douthat interview with Daniel Kokotajlo, executive director of the A.I. Futures Project, in which Kokotajlo suggests that AI could take over civilization—and “then kill all the humans”—by 2027. The existential threat is (potentially) right around the bend.

Maybe. I would suggest, though, that the bigger issue now and for the foreseeable future is AI’s spiritual effects.

In the 19th century, St. John Henry Newman authored An Essay in Aid of a Grammar of Assent. Newman was articulating how the mind moves from doubt or probability to genuine belief, especially in matters of faith. He distinguished between knowledge of abstract propositions (notional assent) and belief that influences behavior (real assent). According to Newman, humans move from notional assent to real assent by relying on what he called the illative sense—a sort of trained intellectual instinct that gathers converging probabilities to make a judgement.1 In the context of religion, the illative sense acts as the “eye of faith,” i.e., how we perceive matters of spirit, and arbitrate matters of soul, when logical inference alone fails to provide certitude.

In considering the relationship between perception, intuition, and assent, Newman demonstrated the “rules” (the “grammar”) by which we come to conviction, belief, and action. Newman didn’t deny the importance of logical deduction, but he situated it alongside this other way of knowing. Belief, or assent, for Newman, is not irrational but suprarational—coinhering but extending beyond mere reason. The precondition for this grammar to “work,” however, is a soul rightly ordered; we can only exercise the illative sense insofar as we assent to God as the unconditioned center of our life. When that center is obscured or ignored, our capacity to perceive and intuit rightly is stunted.

The problem with AI is that it instantiates a grammar of descent.

(Yes, I know that the opposite of assent is dissent, not descent. I’ve chosen this play on words, instead, because, as I describe below, AI technologies tend to pull users into a spiritual descent, regardless of whether the user assents or dissents to divine realities.)

AI-powered social media algorithms provide us with a never-ending stream of curated content that caters to our likes and dislikes; it also gives us a platform to craft and share our ideal self. Streaming media—whether video or music—facilitate me-centered experiences. And now, with generative AI, instead of suffering another person in real and sometimes difficult conversation, I can “converse” with ChatGPT, the partner who only speaks when spoken to, knows everything, and only talks about what I’m interested in. The AI creates a world in which the user is a kind of god, an avatar to which all digital creation must bow.

These technologies are reinforcing and intensifying the postmodern belief that meaning is ultimately derived from one’s subjective experience. In creating and cultivating a technological environment that facilitates hyper-customized digital experiences, we’ve made for ourselves a metaphorical house of mirrors. A house of mirrors is disorienting because it traps us in a self-referential environment. We’re presented with a confounding mosaic of our own reflection, a maze of self-reference that distorts our sense of space and direction.

Analogously, the more we adopt AI-derived habits—that is, the longer we abide in self-referential environments—the more disoriented we become. In the 4th century, St. Augustine described sin as curvatus in se: to be bent (or turned) inward on oneself. It’s a similar image. Artificial intelligence, especially its application in social media and now generative AI, has exacerbated and intensified this tendency toward self-preoccupation; or, better, self-worship.

Whereas in the Holy Trinity Jesus is the source and telos of creation, and we, in Him, are the locus of the Father’s love, the “AI trinity” locates me-the-user as the center of attention. If I can strain the metaphor, instead of the Spirit proceeding as the love between the Father and the Son, the AI algorithm proceeds as the love of the self and for the self.

Artificial intelligence enables and encourages self-worship in another way. The incredible sophistication of AI tools makes us marvel at the work of our own hands. Our belief in the human ethos—specifically, our species’ ability to create—fosters a false sense of independence and self-determination. Even though you or I, and most people, had no direct hand in making ChatGPT, we receive and experience such “gifts” from its human creators in a kind of reverent awe. The effect on individual and collective consciousness is that we become less willing to trust in the divine, and more willing to trust in Man. Whether it’s my own ingenuity or Sam Altman’s doesn’t really matter; it’s “we humans” that created AI, and if we can do that, who needs God?

If the grammar of assent necessitates divine orientation—a relationship with the transcendent that seeks to know beyond what mere reason can deduce—the grammar of descent takes hold as we exclude the divine in pursuit of self-actualization. AI propagates this “language” of self-divination.

In his 2017 bestseller Homo Deus: A Brief History of Tomorrow, Yuval Noah Harari lionizes humans’ capacity for ingenuity and technological augmentation. He writes,

“no god will stop us […] if the whole universe is pegged human experience, what will happen when the human experience is just another designable product, no different in essence from any other item in the supermarket?”2

Such a question assumes a worldview in which, at the ontological level, a person is mere matter. It suggests that human experience is just another “product” to reimagine and recreate, unbeholden to any sort of divine fiat or supernatural order.

The notion that AI advances could conceivably threaten the sanctity of personhood—that such a notion is possible at all—reveals more about sinful human beings than it does about our technological creations. Fallen humanity has proven time and again that we will worship anyone or anything except God. The ancient Israelites crafted and bowed to idols of wood, stone, gold, and other lifeless totems. Modern Americans are quick to worship political parties, sports teams, job titles, or any number of other “gods.” But the danger with AI is that it is so much like us. Generative AI mimics our behavior in such a convincing way that I think it will, and probably already has, exacerbated humans’ tendency to trust in the thing, not in terms of mere utility, but in the existential sense.

If the grammar of descent isn’t rebuffed, we may soon find ourselves worshipping not ourselves, nor our Maker, but the thing we’ve made. That’s the illogic of sin: as we cave in on ourselves, we grasp more feverishly for something to pull us out of the mire of ego. In the “absence” of God, we look to idols, dumb totems that merely refract our sin back onto us.

My concerns are tempered by hope in the Creator who makes all things good, in His time and in His way. In The Meaning of the City, French media ecologist Jacques Ellul struck a hopeful tone about humans’ relationship to technology by appealing to God’s providence. He was writing about the City as an icon of man’s idolatrous pursuit of technological progress, but his claim just as well applies to artificial intelligence:

By this means God gets a foothold in man’s world. He chooses a city, or rather he lets man choose a city for him (after all the city belongs to man!), and by accepting from David’s hands the consecration of man’s counter-creation, God intervenes in the world where man wanted to refuse him entrance. And it is by the hand of man himself that it happens. God does not act as a master able to break down the barriers set up by man, to bring down the walls of Jericho, or to break the gates of Damascus. He does not act as a judge, far above every effort of man to revolt against him […] God meets man on his own ground, on his own terms. As he meets Satan and his spiritual powers where they are.3

In other words, the Cruciform God accounts for all our vainglorious enterprises in His good plan, including our attempts to make idols of ourselves and—in the case of AI—in our own image. The book of Romans puts it more succinctly: All things work to the good for those who love God.4

No matter how far we descend into idolatry, we worship a God who, in Christ, went all the way down for our sake. And He speaks the grammar of ascent.

Footnotes

  1. J. H. Newman (1870). An Essay in Aid of a Grammar of Assent. London, UK: Burns, Pates, and Co.
  2. Y.N. Harari (2016). Homo Deus: A Brief History of Tomorrow. London, UK: Vintage, p. 279.
  3. The Meaning of the City. Grand Rapids, MI: William B. Eerdmans Publishing Co., p. 101.
  4. Romans 8:28

Chase Mitchell

Chase Mitchell is Associate Professor of Media and Communication at East Tennessee State University.

One Comment

  • Gordon Moulden says:

    In both Genesis and Revelation, God is at the center and we are invited in. God is intensely relational and calls us to be likewise, creating Eve not as a sex toy but as a partner for Adam. The two greatest commandments are both relational, and the parable of the Good Samaritan celebrates people who look beyond their own priorities to see the needs of others. Washing the disciples’ feet was relational and commanded we be likewise. AI removes interpersonal relationships and degrades us like few things ever have. Satan is surely smiling.

Leave a Reply