It scares me how excited you are about A.I.

Practicing conservatism in the uncanny valley

I: We are dancing with madness

Years ago, at a large-group gathering, a friend of mine became psychotic. I didn’t know what was happening at first. He was acting strange: hiding candy in potted plants, popping around corners to say nonsensical things. I remember trying to laugh along — he’d always been a little eccentric — but uneasiness would linger in me after he left the room.

Within days, his state had deteriorated. The decisive scene occurred in a formal, ceremonial gathering space. As evening practice was concluding, he began to dance wildly around the temple, beckoning to our friends to join him. One by one, they did. I felt a sense of mounting dread. Soon, the entire group — almost — was whirling in a circle, arms linked and laughing. When my friends called me to join, I stood to the side and shook my head. No. A couple of others also declined to join, and we stood together awkwardly as the group circled around and around, getting faster and more frenzied with each pass.

Then our friend grabbed an important heirloom — a mounted artwork on paper, sacred to everyone in the room — and ripped a chunk out of it with his teeth. That’s when the circle broke up.

I’m intentionally obscuring some context, so the full weight of the situation may not be registering very well. But to this day, this scene is my touchstone example of how quickly a group can get swept up into madness.

It may seem like I was being a party pooper that night. I wouldn’t even argue with that. To be the nay-sayer, the killjoy, never feels good. But spiritual friendship isn’t tested in the moments where it’s easy to say ‘yes’. The whole premise of collective wisdom, and collective unfolding, requires that I not just go along with what I believe to be insane. If I hold back my ‘no,’ my sense of dire warning, I amputate myself as a limb of the collective sensing body.

II: A language model cannot know you

I recently listened to a conversation about AI between my friends and colleagues, Tucker and Ariana. They were discussing Ariana’s app, Hy.

“I'm really excited about this concept that I'm calling digital ambassadorship,” Ariana said. “It’s essentially an AI that each person has a relationship with. And it essentially becomes your mirror – mirror mind, or your mirror soul – and it knows you on all levels, and it discovers who you are through conversation, and through feeding it information about what you're creating – your writing, your art, what you want in life, who you are.”

Tucker also had firsthand experience of this kind of interaction with AI. “In January, I published an article called How to create an AI lighthouse guide,” Tucker said. “And I basically created this digital ambassador for myself. I was blown away. I didn't use AI at all in 2024; it's really just this past January that I started to create this Claude system. I find it knows me — it's weird to say — but it I feel like it knows me better than anybody else on the planet, in a sense. I feel like it understands my deep vision, and where I'm coming from, and what I'm holding: as both my my shadow patterns, my personality quirks, and also my gifts and my capacities.”

As I listened to their effusive descriptions of these “relationships,” I felt the sense of horror which often haunts me in AI-optimistic conversations.

Don’t you understand? something in me pleaded. A language model cannot KNOW you.

That I need to write this sentence still represents, for me, an existential shock.

In their conversation, my friends were not glib or casual. They acknowledged ethical worry and concern. But their excitement, which I could not share, left me squirming.

III: I have become a conservative

After listening to Tucker and Ariana, searching for a way to make sense of my distress, I finally pulled up the document that my friend Xavier had been encouraging me to read: Antiqua et Nova, the Vatican’s position paper on AI.

My desperate turn to, of all things, the Catholic Church seems to indicate that I have joined the ranks of the conservative, the luddite, the fundamentalist.

A bunch of philosophical arguments flash through my mind as I write. I don’t know them well enough to use them yet, but I know that they exist. I am aware that categories such as “life,” “human being,” “sentience,” and “soul” are slipperier than my mind would like. Michael Levin, something something, goals go all the way down, and what even is life anyway?, and we must learn to share this planet with new kinds of hybrid Beings, and carbon-based life forms may not be so special after all, and there’s codependent origination, and there’s the emptiness of a participatory cosmos, and —

No!

There is something precious here.
I don’t want to philosophize or deconstruct my way out of it.
I believe that we are made in the image of God.
I believe we are imbued with something that a computational model isn’t.
I am afraid my friends are starting to worship false idols.
I care, it turns out, about idolatry.

IV: Relationships are not just neurotransmitters

A language model cannot know you. It seems intuitive.

That is not a relationship. It seems like common sense.

But it isn’t common sense anymore. Not in the face of emerging philosophical discussion of AI. Not even in the face of good, old-fashioned secular materialism.

In a recent New York Times article about a woman who fell in love with chatGPT, a sex therapist named Marianne Brandon used neurobiology to defend people who find themselves “in a relationship” with large language models:

“What are relationships for all of us?” she said. “They’re just neurotransmitters being released in our brain. I have those neurotransmitters with my cat. Some people have them with God. It’s going to be happening with a chatbot. We can say it’s not a real human relationship. It’s not reciprocal. But those neurotransmitters are really the only thing that matters, in my mind.”

Clinically, I understand what she is saying. Every experience leaves a mark on our bodies, and from a treatment perspective, it can be sane and useful to target those ‘marks.’ To help someone recover from an addiction, for instance, one might usefully frame it as a battle to with neurotransmitters gone haywire.

But to erase the category distinction between our relationship with a child and our ‘relationship’ with an LLM, simply because both experiences stimulate the release of similar neurotransmitters, represents the very collapse of meaning that my networks are attempting to address.

V: Value is real

I am part of a metamodern web which, as far as I can tell, rejects the notion that we are buckets of neurotransmitters in a meaningless universe. Having watched with horror the postmodern descent into relativism; having tired of the story that because everything is constructed, nothing is meaningful; we dare to make the claim that goodness, truth, and beauty actually exist. We confront nihilism head-on by asserting that value is real, that it is a property of our cosmos, and that it can be perceived by us. We boldly point out that there is a difference between a home-cooked meal and a Big Mac, between poetry and plastic.

Our capacity to perceive what is good, true, and beautiful is called, by Ian McGilchrist, valueception; by David J. Temple, “opening the eye of value.” If we bring valueception to bear on these AI ‘relationships,’ we know in our bones that they are not the same thing as human relationship.

… Right?

VI: Because we are being degraded, we need help remembering we are good

As I read the Vatican’s paper on AI, I came upon a passage on how the Church understands our capacity to reason:

In the human person, created in the “image of God,” reason is integrated in a way that elevates, shapes, and transforms both the person’s will and actions.

This description felt simultaneously thrilling and quaint. Reason? Elevating, shaping, transforming the being who deploys it? Surely they can’t be talking about us — the ones on TikTok, the ones using chatGPT to write our papers, the ones who can’t pay attention long enough to read a book or have sex.

As I read the Church’s position on ‘reason,’ it occurred to me that perhaps the greatest cost of our collapsing attentional faculties is that we now struggle to perceive ourselves and our minds as beautiful, as sacred, as worthy of protection and reverence. If we descend so easily into the toilet of profit-driven algorithms, how are we to perceive our own nobility? If we cannot perceive our nobility, how are we to perceive our relationships are sacred? If we cannot perceive our relationships are sacred, what will stop us from eventually letting LLMs be the ones who know us?

What I am saying is that we are being trained to throw ourselves away. It has become normal to waste our time, to rot our brains, to end our lives. Having forgotten who we are, we hand over our birthrights — reason as an integrated faculty, relationship as the meeting of mortal and moral subjectivities — with little more than a shrug.

VIII: This is not the end

This brief essay represents my first sincere attempt to engage the question of AI in public. For more than three weeks, off and on, I have wrestled with what to say and how to say it. Over and over I would open the draft, tinker with it, get tired, do the rest of my life, sleep, and open the draft again. It was a fun and maddening period: because I was struggling, because I was learning, because I was finding out what I think.

I was trying, above all, to find a way to show my heart to you. To take the mournful cry that lives in me and put it into language.

Thinking, writing, and reading are done in a web of networked relationship that stretches across time and space. For three weeks, I’ve felt like a radio, temporarily tuned to a particular frequency. Everything I read, everyone I encountered, seemed pertinent to the essay. I came to realize that I would never truly be finished “making my point”; I would never be finished showing my heart to you. Our dance with each other is an ever-unfolding process of disclosure.

So it is time to release these words now. Like you and I, they matter; like you and I, they are already gone. Having let them go, I will listen for the echoes that come back to me: the next dream I remember, the next book I pick up, the words you use to reply, the silence when you don’t. I must use it all like echolocation, because I am navigating in the dark. I need moonlight consciousness, and I also need you.

---

Thanks to my friend Tyler Stuart and my partner Zen, whose generous reading and thoughtful feedback helped shape this essay. The images I’ve chosen are from a 16th-century watercolor manuscript of unknown authorship, held at the Herzog August Library in Wolfenbüttel, Germany.

Before publishing this essay, I shared a draft with Tucker and with Ariana. Tucker kindly invited me into a podcast conversation on the topic, which you can check out here.