For almost six years, I’ve been contributing to Standard Ebooks, a volunteer project which (in our own words) “produces new editions of public domain ebooks that are lovingly formatted, open source, free of U.S. copyright restrictions, and free of cost.” By my count, I’ve produced nearly fifty ebooks over the years, including both heavy-hitters like The Idiot and much more obscure titles like Ameen Rihani’s poetry.1 These days you will find my photo on our masthead, because I am a member of the editorial team, which is responsible for overseeing the projects of other contributors.

My work as a volunteer is time-consuming. I would guess that on a given week, I spend between five and ten hours on the site, whether I am working on my own ebook projects or providing feedback on others’ projects. But my work is freely chosen and freely given, and at this point, I could not imagine my life without it. I do not and will not regret a single second of my life that goes into making a beautiful ebook.

In fact, I started contributing to Standard Ebooks at a difficult period in my life. During the fall of 2020, about half a year into the Covid pandemic, I was struggling to make sense of what I was doing with my life as I was holed up alone in my studio apartment in Palo Alto, working remotely in my software engineering role. Day in and day out, I dragged myself to my laptop to do my work and make my paycheck, but why? Why was I working so hard to get by in a world that felt so stifling and bleak?

It turned out that the answer was Leo Tolstoy. Once when I was young and still in middle school, I stumbled upon his short story, “How Much Land Does a Man Need?” It was the first time a piece of classic literature really captured my attention, and I never quite forgot about it because years later, in 2019, it somehow occured to me that I wanted to read the story again and rekindled my interest in Tolstoy’s stories. To a great extent, I credit Tolstoy with my conversion to Christianity. I am not sure if I would have ever embraced my Christian faith without the help of Tolstoy’s short stories, which have done much to help me see the beauty of loving one’s neighbor.

So, after I started contributing to Standard Ebooks, it felt very natural for me to work on a digital anthology of all of Tolstoy’s short stories.2 When I began the project, I had little idea of just how much work the project would actually involve. There are about a hundred English translations of Leo Tolstoy’s short stories which have gone out of copyright in the United States, and together they add up to about 650,000 words, which is almost a hundred thousand words longer than the entirety of War and Peace.

You might be wondering how much time this took. The truth is, I don’t really know myself, but in some sense it was less than you might have imagined. I finished the collection in about half a year, because I was putting a considerable portion of my free time into the project. The pandemic gave me very little else to do and the project gave me a sense of being part of something much larger than myself. Years later, I still consider it to be one of my life’s greatest achievements. As far I know, it was the first time anyone compiled all of Tolstoy’s short stories into a free and open source ebook.

Now, early on during my time as a Standard Ebooks contributor, it occured to me that, on some level, the work we were doing was not that complicated and that in the nearish future, it might be possible for an AI system to do our work from end to end. ChatGPT had not yet come out yet, but I was vaguely aware of LLMs because I had spent some time playing around with AI Dungeon, a text-based game built on GPT-2, back when OpenAI was an obscure nonprofit better known for training Dota 2 bots.

I have never used Claude Code or any other AI coding harness before, but I suspect that with the right configuration (i.e. creating the right skills), it should be possible to have Claude Code produce an ebook from end to end with minimal human intervention. There are some aspects of producing an ebook, like choosing a cover and writing a description that can’t (or shouldn’t) be delegated to AI, but in principle, I would guess that I could get Claude Code to do at least 75% of my development work, in such a way that would have little to no noticeable impact on code quality. I suspect that we are not that far from AI models that could bring that number to 100%.3

I’m writing this because although I have a generally critical attitude towards generative AI and the AI industry, I want to acknowledge that it can be genuinely useful for a lot of things, and there are many interesting ethical questions that emerge out of that usefulness. If Claude Code is as good at helping me develop ebooks as I think it might be, and if I think that developing free and open source ebooks is an unqualified good, should I be using Claude Code to speed up my ebook development?

Now, I could make some argument that regardless of how helpful Claude Code might be in this case, I should not be giving Anthropic business for other reasons. Indeed, as I highlighted in “You can’t just kill people,” Anthropic has expressed an openness towards the future development and utilization of "fully autonomous” weapons, a position which I think is morally abhorent on a fundamental level. But let’s suppose, for sake of argument, that this wasn’t a problem, and, in fact, we lived in a magical world where AI used no physical resources like water or power. What then?

This is the question I am trying to get at: What is the point of coding ebooks? Or, if I want to get a little Aristotelian: What is the telos (that is, the end, the final purpose) of coding ebooks? Hence, I quote from the opening of Aristotle’s Nicomachean Ethics:

Every craft and every line of inquiry, and likewise every action and decision, seems to seek some good; that is why people were right to describe the good as what everything seeks. But the ends that are sought appear to differ; some are activities, and others are products apart from the activities. Whenever there are ends apart from the actions, the products are by nature better than the activies.4

Is ebook development a craft, whose goal, the production of ebooks, is external to itself? Or, is ebook development something that is worth doing for its own sake?

Thinking about the issue in terms of a strict binary is unlikely to help. If ebook development is what the moral philosopher Alasdair MacIntyre called a “practice,”5 it involves shaping me to be a certain kind of person with particular standards of excellence, even as I am involved in the production of digital goods. Ebook production involves some labor-intensive manual steps like adding semantic information and completing an end-to-end proofread. Although on some level the purpose of these tasks is to make the final product come out a certain way, I find that they also fulfill a deeper purpose of deepening my habits of attention and respect for others’ work.

You have to be a certain kind of person to be willing to proofread an entire novel to iron out all of the little typographical errors you find (believe me, there can be many). Also, now that I know how much work goes into publishing an ebook even with the help of cutting-edge automation tools, I have a much greater appreciation for the work that was involved in the original physical publication of these books. If proofreading and making typographical corrections is so frustrating when I am working with digital files, I can only imagine what it was like to work with a physical press!

It reminds me of a vision by the Russian revolutionary Peter Kropotkin, that some day the producers of academic knowledge would not snub their noses at the manual labor involved in publishing their journals, but would take part in the process themselves:

It is true that a learned society, like the individual author, goes to a printing office where workmen are engaged to do the printing. Nowadays, those who belong to the learned societies despise manual labour which indeed is carried on under very bad conditions; but a community which would give a generous philosophic and scientific education to all its members, would know how to organize manual labour in such a way that it would be the pride of humanity. Its learned societies would become associations of explorers, lovers of science, and workers⁠—all knowing a manual trade and all interested in science. … In future, when a man will have something useful to say⁠—a word that goes beyond the thoughts of his century, he will not have to look for an editor who might advance the necessary capital. He will look for collaborators among those who know the printing trade, and who approve the idea of his new work. Together they will publish the new book or journal.6

By engaging in particular practices (in my case, ebook production), we create our conceptions of who we are: “a concept of a self whose unity resides in the unity of a narrative which links birth to life to death as narrative beginning to middle to end.”7 In my life, the practice of ebook production has been part of a narrative of discovering spirituality and creating meaning, from my childhood encounter with Tolstoy to the soul-crushing months of the Covid-19 pandemic. This narrative is not something I created from whole cloth, for myself, but through encountering other narratives: for instance, Leo Tolstoy’s short stories, or the sacred stories of the Christian canon.

But as the philosophers of technology Wessel Reijers and Mark Coeckelbergh argue, however, narratives are created through our encounters with technology, as much as they are through our encounters with literary texts.8 When I use the Standard Ebooks toolset, I develop (whether consciously or not) a plot in my head, something like “I’m using cutting-edge tools in order to produce beautiful and accessible ebooks.” But if I started farming out my work to Claude Code, what other plot might I be forming? “I’ve reviewing the work of an AI agent because this will produce faster output”? Which of these two plots would I rather live under? Which of these plots will help me become the sort of person I want to be: someone who deeply appreciates literature?

In writing this essay, I don’t want to speak for everyone, or insinuate that it is impossible to use AI tools in a way that is compatible with deeper values. But if we reach for these tools by default because we have been shaped by a cultural paradigm of efficiency (what Jacques Ellul called technique in The Technological Society), we will never even get to the point of making that reflection in the first place. This is, I think, the important contribution of virtue ethics to the field of AI ethics. We can’t judge whether we are using generative AI in an appropriate way without first reflecting on the kinds of people we want to be, and whether generative AI is compatible with that.

I adapted the title of this essay, “Work can be joyous,” from the short story “Walk in the Light While There Is Light” by Leo Tolstoy. I can’t entirely explain the title without spoiling the story’s plot, so I invite you to read the story for yourself. But I want to leave my readers with this question: What kinds of work are good to do for their own sake? What is something I would never want to automate out?

Earlier this month, on March 4, 2026, The Washington Post published an article highlighting the ways in which Claude, Anthropic’s generative AI model, has been used by the U.S. military to wage war against Iran. A brief excerpt from the article:

As planning for a potential strike in Iran was underway, Maven [a system built by Palantir], powered by Claude, suggested hundreds of targets, issued precise location coordinates, and prioritized those targets according to importance, said two of the people. The pairing of Maven and Claude has created a tool that is speeding the pace of the campaign, reducing Iran’s ability to counterstrike and turning weeks-long battle planning into real-time operations, said one of the people. The AI tools also evaluate a strike after it is initiated, the person said.

Evidence suggests that one of the targets struck by the United States military on the first day of the war was the Shajarah Tayyebeh elementary school, which was located next to an Iranian naval base. It is reported that the strike killed at least 175 people.

A few days before the war against Iran began, Pete Hegseth, U.S. Secretary of Defense, demanded “full, unrestricted access to Anthropic’s models for every LAWFUL purpose in defense of the Republic,” or else he would punish Anthropic by designating it as a supply-chain risk, preventing Anthropic’s models from being used in the military’s suppliers. What kind of “full, unrestricted access” was Hegseth so interested in obtaining? According to a statement from Anthropic, they refused to include two use cases in their contracts with the Department of Defense: mass domestic surveillance, and fully autonomous weapons. Since then Anthropic has seen a surge in app downloads while Hegseth has carried out his threat, leading to lawsuits.

I am writing about these events in my newsletter, not because I think the Internet needs another piece of punditry about a brutal act of violence against children, but because as a theologian of technology, I see these events as a dire warning that our current technological trajectory will lead us towards complete moral bankruptcy.

On some level, I think it’s commendable that Anthropic stuck by its hard lines. It would certainly be abominable if the Department of Defense was granted free reign to militarize Anthropic’s models and jam them into fully autonomous weapons. But a close reading of Anthropic’s statement suggests that the gap between what Pete Hegseth wants, and what Anthropic is willing to provide, is not that large:

Partially autonomous weapons, like those used today in Ukraine, are vital to the defense of democracy. Even fully autonomous weapons (those that take humans out of the loop entirely and automate selecting and engaging targets) may prove critical for our national defense. But today, frontier AI systems are simply not reliable enough to power fully autonomous weapons. We will not knowingly provide a product that puts America’s warfighters and civilians at risk. We have offered to work directly with the Department of War on R&D to improve the reliability of these systems, but they have not accepted this offer.

It is not the case that Anthropic has any philosophical objection to fully autonomous weapons. Rather, their objection is that they cannot make good enough weapons currently! From this point of view, the futures that Anthropic and the Department of Defense envision are not so different. Both want to increase the abstraction of war, putting more and more layers between the act of violence and the responsible human.

In Narrative and Technological Ethics, Wessel Reijers and Mark Coeckelbergh observe that sociotechnical systems shape our worldview because they abstract us away from the world of action, where actual, specfic people do actual, specific things. A sociotechnical system becomes a “quasi-entity,” something that we assign agency to in our narratives, even though agency ultimately belongs to particular human beings.1

Looking back at the quote from The Washington Post which I used to open this article, we see this kind of abstraction in full force in the very first sentence. “Maven, powered by Claude, suggested hundreds of targets, issued precise location coordinates, and prioritized those targets according to importance.” Does it really make sense to say that “Maven, powered by Claude” suggested targets? This kind of anthropomorphic language only makes sense if you were to think about Maven or Claude as though they were concrete entities with actual agency, which they decidedly are not.

It would be like if I flipped a coin to decide whether or not I should kill someone, and after killing them, I declared, “The coin decided to kill them!” Of course these AI systems are much more sophisticated than a coin flip, but the principle is the same. Claude doesn’t suggest targets; my coin doesn’t decide to kill people; it is humans who do those things. And from this point of view, the entire idea of a “fully autonomous weapons system” is simply a misnomer. There is no such thing because some number of actual human beings set the weapons system into motion, intending violence.

What I find most troubling in this pattern of increasing abstraction in war is that it will fundamentally close us off from our basic moral capacities as human beings. In the Gospel of Mark, Jesus opens his ministry by declaring, “The time is fulfilled, and the kingdom of God has come near; repent, and believe in the good news.”2 The Greek word translated here as “repent,” metanoeite, means something like “change your minds.” How are people going to change their minds when increasing layers of technical abstraction cause people to hide their own actions from themselves?

Even the image of Pilate washing his hands3 doesn’t quite capture the pattern I am concerned about. In the Gospel narrative, Pilate knew fully well what he was doing, but denied responsibility for his actions. What I am contemplating is a world in which people set up “fully autonomous” weapons systems thinking to themselves in all sincerity that the weapons are “deciding” who to kill, when at base it is the humans who are responsible. If this does not seem entirely believable to you, consider this: AI tools are already being used to abstract away the act of denying people healthcare. Is using AI to abstract away the act of killing people really that big of a next step?

One of the slogans that the cultural apologists of Silicon Valley love throwing around these days is the idea that “you can just do things.” Case in point:

It is, indeed, much easier to do things if you don’t know what you are really doing! The same is true of killing people. You can’t just kill people. You can’t just do things!

I had some second thoughts about giving this essay on advertising in AI platforms the title that I did. AI platforms are many things, but I would not include “My Father’s house”1 in that list. But my reference does capture an important aspect of the Gospel text: the sense of moral indigation I feel when I think about ads in AI platforms!

In this month’s essay, I will lay out, within a framework of theological virtue ethics, why I find ads in AI platforms so ethically disturbing. (You will quickly see that many of the arguments I make here apply to many other forms of apps and websites.)

Virtue ethics is a philosophical approach within ethics that evaluates the merits and demerits of various actions in terms of how they form a person’s character. A virtue is an excellent character trait: “a disposition well entrenched in its possessor—something that, as we say, goes all the way down, unlike a habit such as being a tea-drinker—to notice, expect, value, feel, desire, choose, act, and react in certain ways.”2 Other major ethical approaches include consequentialism,3 which emphasizes the consequences of one’s actions, and deontology, which emphasizes moral duties.

Virtue ethics has been articulated in various forms by the world’s major wisdom traditions. To offer just a few examples, we might think about Plato and Aristotle in the West, or Confucius in China. In the Christian theological tradition, St. Thomas Aquinas elaborated on Aristotle’s virtue ethics in the Summa Theologica. Following earlier philosophical traditions, he enumerated prudence, justice, temperance, and fortitude as the cardinal virtues, which encompass all other human virtues.4

We might ask, at this point, why we consider these character traits to be so excellent. From a Christian theological standpoint, I suggest looking towards Genesis 1:26–27, which is the Scriptural basis of the imago Dei—the idea that humans are made in God’s image. As I explained in my essay “AI, humanity, and the image of God,” there are three major approaches to understanding the imago Dei, which each describe different kinds of excellent activities by which humans express their resemblance to God.

Humans don’t become faithful images of God through sheer happenstance. Rather, there are certain character traits that support the activities God calls us to do. In this article, I am most interested in analyzing virtue in terms of the functional model of the imago Dei. This model builds on the Biblical language of humans having “dominion” over the animals, suggesting that humans are in God’s image because they serve as representatives of God’s reign over creation. In other words, humans serve as stewards over creation, maintaining the world in good order on God’s behalf. We might think about this in environmental terms: have we humans taken good care of nature?

Therefore, when I analyze AI from a theological ethics perspective, my first question is, “Will this development reinforce, or undermine, the virtues by which people become faithful images of God?” My question builds on the observation by the philosopher of technology Shannon Vallor (written in the context of discussing carebot ethics) that analyses of human-machine interaction are “incomplete until we have also considered the possible impact … on human habits, virtues, and vices.”5 I am especially concerned about the impacts of LLMs on human character formation, because LLMs provide a simulacrum of social interaction, and we have known (from the time of Aristotle!) that repeated social interactions form human virtues and vices.

So what virtues might undergird the functional model of the imago Dei—that is to say, make us more faithful stewards of God’s creation? I think that the most relevant cardinal virtues are those of prudence, and temperance. Prudence (in Greek, phronēsis), is the virtue of “right reason applied to action,” which enables humans (as finite beings) to make good judgments about the particular situations they will encounter.6 Temperance (sōphrosunē) is a virtue that governs our relation to sensual pleasures which arise from life-sustaining human activities (e.g. eating and drinking, having sex).7 Our ability to function well as stewards of creation depends on both of these virtues: we must be able to make good judgments about how our actions will affect the environment, and we must govern our own appetites to avoid wasteful consumption.

So how does advertising come into this? Before I get into the ethics of advertising on AI platforms, I want to explore the ethics of advertising in general. Roger Crisp’s article “Persuasive Advertising, Autonomy, and the Creation of Desire”8 lays out a systematic critique of persuasive advertising, and is well worth reading. Persuasive advertising seeks to create desires on a subconscious level, using techniques like subliminal messaging, sex appeal, and repetition. This is in contrast to informative advertising, which (in principle) seeks to provide potential customers with relevant facts about the product to inform their decision-making. As an example of persuasive advertising, we might consider this infamous ad for HeadOn from nearly twenty years ago. What is HeadOn? Why apply it directly to the forehead? God only knows!

Crisp argues that persuasive advertising is ethically questionable because it causes desires to arise without ever allowing the consumer to make a conscious decision, thus overriding their autonomy. From a Christian virtue ethics perspective, I see persuasive advertising as being problematic on two accounts. First, people cannot be prudent (i.e. acting according to the right reason) if they are not consciously acting according to any reason at all. Secondly, people have a much harder time being temperate if they are constantly being bombarded with subliminal appeals to their appetites. Even before generative AI, we have seen how social media recommender systems have given rise to the influencer economy and digital platforms for pornography like OnlyFans.

So now I turn to the issue of advertising in AI platforms, the subject of this essay. You might have heard that about a month ago, OpenAI announced plans to roll out advertising in ChatGPT. On first glance, their mockups don’t look too terrible. The ads are separated from the ChatGPT response, and seem informative in nature. But in practice, I argue that any advertising in platforms like ChatGPT is inherently manipulative, because of the ways LLMs are optimized to maximize engagement.

Large language models (LLMs) like GPT-5, which powers ChatGPT, are optimized as chatbots through the process of reinforcement learning from human feedback (RLHF).9 As Sharma et al. have observed, one of the emergent behaviors of LLMs finetuned through RLHF is sycophancy, in which LLMs tend to generate output that agrees with whatever the user’s views and biases are.10 It turns out that one of the most powerful factors for predicting whether people will find a given LLM output helpful is whether or not that output agrees with what they already believed! To be clear, I don’t think that AI researchers ever intentionally sought to develop sycophantic systems, but in retrospect this outcome doesn’t seem very surprising. A system trained to capture human preferences will also capture unconscious biases!

I argue that the effectiveness of advertising is fundamentally tied to sycophancy, because in many cases this behavior encourages repeated usage of LLMs. For example, just think about how OpenAI brought back the infamously sycophantic model GPT-4o after loud complaints from the userbase. Fundamentally, advertising on a LLM platform is effective only because RLHF captures unconscious patterns of human behavior, which can be used to manipulate users into repeated usage and even (in extreme cases) emotional dependence. Therefore I strongly assert that it is unethical for any AI platform to serve advertisements alongside a model finetuned using RLHF.

[ Center for Humane Technology ]

Advertising is Coming to AI. It’s Going to Be a Disaster.

This piece was published in Tech Policy Press on Nov 27, 2025, and has been republished with permission…

Read more

5 months ago · 241 likes · 34 comments · Daniel Barcay and Center for Humane Technology

I am heartened that , with the , has put out a prescient warning about these very issues linked above. I wrote the essay you are reading as part of a conference paper, without being aware of Barcay’s article. It’s very good and I would gladly second all of their recommendations about guardrails.

Turning back to my original comments on this essay’s title, I realized that the admonition “make not my Father’s house a house of merchandise” is relevant in a surprising way. Obviously, AI platforms are not “my Father’s house,” but as Paul wrote to the Corinthians, “Do you not know that you are God’s temple and that God’s Spirit dwells in you? If anyone destroys God’s temple, God will destroy that person. For God’s temple is holy, and you are that temple.”11 Our very bodies and minds are temples and houses of worship. Let the AI companies hear Christ’s proclamation: “Take these things hence; make not my Father’s house a house of merchandise!”

It wasn’t until about a month ago, when I read about the history of Herod’s Temple in preparation for a sermon, that I had any idea of its scale and opulence. I came across a passage by the Roman-Jewish historian Josephus describing the Temple’s facade, which I find immensely helpful for visualizing what Jesus was speaking about:

Now the outward face of the temple in its front wanted nothing that was likely to surprise either men’s minds or their eyes; for it was covered all over with plates of gold of great weight, and at the first rising of the sun, reflected back a very fiery splendour, and made those who forced themselves to look upon it, to turn their eyes away, just as they would have done at the sun’s own rays. But this temple appeared to strangers, when they were coming to it at a distance, like a mountain covered with snow; for, as to those parts of it that were not gilt, they were exceedingly white. On its top it had spikes with very sharp points to prevent any pollution of it by birds sitting upon it. Of its stones some of them were forty-five cubits [~20 m] in length, five [~2.3 m] in height, and six [~2.7 m] in breadth.1

Seeing it put this way, it doesn’t seem so surprising that the people around Jesus were so caught up in admiring the architecture. Herod’s building project seems to have had its desired effect, being so visually astonishing that people are distracted from seeing how God is at work in the world. I should point out that just before this exchange, Jesus highlighted the wealth disparities of ancient Judea by contrasting a poor widow who put two copper coins into the temple treasury with the rich who put in their own (presumably lavish) gifts. But the people would rather talk about beautiful stones!

Then I ask myself, who are we to judge? Aren’t the people of our generation caught up in admiring the works of human hands? Gold and stones; machinery and circuits; isn’t it all the same? Will the day not come once again when “all will be thrown down”?

Text within this block will maintain its original spacing when published

For the Lᴏʀᴅ of hosts has a day
 against all that is proud and lofty,
 against all that is lifted up and high;
against all the cedars of Lebanon,
 lofty and lifted up;
 and against all the oaks of Bashan;
against all the high mountains
 and against all the lofty hills;
against every high tower
 and against every fortified wall;
against all the ships of Tarshish
 and against all the highly prized vessels. —Isaiah 2:12–16

The stones of Herod’s Temple were certainly durable; left alone, they might have endured for generations. From a worldly point of view, it is understandable that people would trust in stones and machines, things that do not die—or, for that matter, live. But the durability of stone belies its vulnerability. Stones and the things we build from them are not eternal. They are subject to time and chance, just as we are.

The people of ancient Judea learned this lesson the hard way when, in the aftermath of the First Jewish-Roman War, the Romans destroyed the Temple and looted its precious treasures, including the Menorah. The four Gospels, written in the years and decades after the Temple’s destruction, speak to the lessons of this bitter experience:

The Passover of the Jews was near, and Jesus went up to Jerusalem. In the temple he found people selling cattle, sheep, and doves and the money changers seated at their tables. Making a whip of cords, he drove all of them out of the temple, with the sheep and the cattle. He also poured out the coins of the money changers and overturned their tables. He told those who were selling the doves, “Take these things out of here! Stop making my Father’s house a marketplace!” His disciples remembered that it was written, “Zeal for your house will consume me.” The Jews then said to him, “What sign can you show us for doing this?” Jesus answered them, “Destroy this temple, and in three days I will raise it up.” The Jews then said, “This temple has been under construction for forty-six years, and will you raise it up in three days?” But he was speaking of the temple of his body. —John 2:13–20

God’s temple is not made of lifeless stone, but human flesh. God’s temple lives, and moves, and breathes, and suffers, and dies, in order to live again. From a worldly point of view, it seems that stones will outlast flesh. But we know very well that although the stones are enduring, they are not eternal! If we want to hope in something eternal, we must not look towards “houses made with human hands,”2 but instead towards God’s promise of the resurrection, sealed by the One who is true God and eternal life.3

Text within this block will maintain its original spacing when published

Do not court death by the error of your life
or bring on destruction by the works of your hands,
because God did not make death,
and he does not delight in the destruction of the living. —Wisdom 1:12–13

But in our day and age, and perhaps in every day and age, those who hold the levers of economic and political power grasp for the secrets of immortality in the most absurd places, instead of trusting in the promise of God who is Faithful and True.4 Just the other day I read ’s latest dispatch from Silicon Valley, highlighting how (I wish I was joking) the tech bros of Silicon Valley are injecting themselves with grey market Chinese peptides and resurrecting Victorian racialist pseudoscience. The ridiculousness of it, I think, slightly undercuts how abominable it actually is: to treat yourself like a lab rat, and your body like a machine, and for what purpose? To stave off, for a few years, a bodily death which must surely come? What madness!

And this is just the tip of the iceberg. As points out, the worlds of transhumanism and Christian nationalism are increasingly coming together. On first glance this might seem surprising, but on closer inspection we see that these ideologies have deep structural similarities: eschatological promises of weal and woe; the idolatrous exaltation of human power in the form of machines and nations. Some transhumanists even refer to a theoretical artificial superintelligence (ASI)—an AI whose general capabilities greatly exceed those of humans—as a godlike being.

But God does not dwell in houses made by human hands, whether those houses are made of stones, or of semiconductors! God dwells in each and every one of us. For as St. Paul wrote to the Corinthians, “Do you not know that you are God’s temple and that God’s Spirit dwells in you? If anyone destroys God’s temple, God will destroy that person. For God’s temple is holy, and you are that temple.”5 What is the good news but this: that just as God raised the temple of Jesus’ body from the dead, we too are temples of God, and God will raise each of us from the dead in the fullness of time?

So in this new year, let us ask ourselves, where shall we direct our efforts? Let us “build our hopes on things eternal,” remembering that of all things, only love never ends.6 And when all is said and done God will not ask us whether we spent enough time admiring the architecture, but whether we have fed the hungry, quenched the thirsty, welcomed the stranger, clothed the naked, and visited the prisoner.7 Shall we not be faithful and take up God’s call to have mercy when the ruling powers will not?

Text within this block will maintain its original spacing when published

Time is filled with swift transition,
naught of earth unmoved can stand;
Build your hopes on things eternal,
hold to God’s unchanging hand!

Thank you for reading my first post of the year. I’m hoping to keep my publishing streak going throughout the year. Please consider sharing my work with anyone who might want to read it!

Share

One day as we were going to the place of prayer, we met a female slave who had a spirit of divination and brought her owners a great deal of money by fortune-telling. While she followed Paul and us, she would cry out, “These men are slaves of the Most High God, who proclaim to you the way of salvation.” She kept doing this for many days. But Paul, very much annoyed, turned and said to the spirit, “I order you in the name of Jesus Christ to come out of her.” And it came out that very hour.

But when her owners saw that their hope of making money was gone, they seized Paul and Silas and dragged them into the marketplace before the authorities. When they had brought them before the magistrates, they said, “These men, these Jews, are disturbing our city and are advocating customs that are not lawful for us, being Romans, to adopt or observe.” The crowd joined in attacking them, and the magistrates had them stripped of their clothing and ordered them to be beaten with rods. After they had given them a severe flogging, they threw them into prison and ordered the jailer to keep them securely. Following these instructions, he put them in the innermost cell and fastened their feet in the stocks. —Acts 16:16-24

When it comes to interpreting the Bible, I think we are often tempted to boil down a complex passage into a pithy moral. That being said, I think that some amount of simplification is inevitable and indeed desirable. If a Christian really wishes to live, not “by bread alone, but by every word that comes from the mouth of God,”1 it is necessary to search Scripture for actionable principles to live by. I think that God shows us grace in the fact that we can be edified by our reading of the Scriptures, even when we don’t understand precisely what the author originally intended. We would be in a truly sad state if our spiritual renewal depended solely on our exegetical prowess. The key is that we must read Scripture to build up love. For as St. Augustine wrote:

Whoever takes another meaning out of Scripture than the writer intended, goes astray, but not through any falsehood in Scripture. Nevertheless, as I was going to say, if his mistaken interpretation tends to build up love, which is the end of the commandment, he goes astray in much the same way as a man who by mistake quits the high road, but yet reaches through the fields the same place to which the road leads. He is to be corrected, however, and to be shown how much better it is not to quit the straight road, let, if he get into a habit of going astray, he may sometimes take cross roads, or even go in the wrong direction altogether.2

When it comes to interpreting Acts 16:16–24 for this month’s essay (really, it’s more a homily), I fear that a detour through the fields might be inevitable. It is hard for me to guess precisely what the Evangelist intended to convey with this passage, but I can’t imagine that he had the potential perils of AI chatbots in mind. Without further explanation, that interpretation may seem like something of a stretch, but I will try to show that there is a surprising structural parallel between the case of the enslaved woman and the recent proliferation of AI chatbots used for spiritual companionship.

When it comes to interpreting this passage, I think that the pivotal question is “What motivated Paul to exorcise the enslaved woman?” If we were only to look at what the woman’s public utterances—“These men are slaves of the Most High God, who proclaim to you the way of salvation”—the problem does not seem so obvious. There is nothing in her words that contradicts how Paul portrays himself, whether in the narratives of Acts or in his own letters. In the first verse of Romans, Paul identifies himself as a slave (in Greek, doulos) of Jesus Christ. A few verses later, in Romans 1:16, Paul declares that the gospel is “God’s saving power for everyone who believes.” We also know that early Christians were called “the Way.” Everything seems to check out!

The issue is that if we limit ourselves to an analysis of the woman’s speech taken out of context, we are engaging in what an anthropologist might call “thin description.” While our thin description captures behavior on the surface level (i.e. what the woman says), it does not capture the woman’s intentions or her social world. For that, we need to engage in “thick description,” interpreting her behavior within her context.3

In order to provide a thick description of these events, we need to make some important observations about the woman’s social location. The passage describes her as “a female slave who had a spirit of divination and brought her owners a great deal of money by fortune-telling,” which provides us with several relevant details.

First, as a female slave in the Roman Empire, this woman was constantly surrounded by the threat of legally-sanctioned violence. Under Roman culture, enslaved people were viewed as extensions of their enslavers’ bodies, such that the Greek word for body, sōma, came to be used as an euphemism for “slave.” (In fact, in Revelation 18:13, merchants lament that with the fall of Babylon, they no longer have a market for “bodies,” which is to say, human lives!) Under Roman law, all freeborn people had license to insult enslaved people, while enslavers could subject their slaves to physical torture.4 Slaves were subjected to pervasive sexual violence, because the law granted enslavers unrestricted sexual access to their slaves. Female slaves were used as sexual surrogates that would engage in acts that freeborn women considered shameful.5

Furthermore, the text tells us that the woman had a spirit of divination. While women in the Roman world were generally expected to keep silence, one of the accepted avenues for women to speak was through prophecy. For example, the Oracle at Delphi, also known as the Pythia, was thought to deliver prophecies from Apollo.6 This woman’s prophetic possession provided her a socially-acceptable avenue for public speech, which, as an enslaved woman, she might not have had possessed otherwise. However, the text informs us that the woman did not speak purely for her own benefit, because she brought her owners a great deal of money by fortune-telling. This woman could not speak freely, because as an enslaved person in the Roman Empire, she had no legal protections from being tortured or sexually coerced by her enslavers.

Therefore, when we think about Paul’s act of exorcism in this passage, we cannot think about it simply in terms of what the enslaved woman is saying, but we must also consider the fact that ultimately, her speech is controlled by her enslavers. Although the woman is physically capable of all kinds of speech, she is socially constrained by her enslaved status and the omnipresent threat of her enslavers’ violence.

I think these facts are important in light of the fact that she kept doing this for many days. Her utterance, “These men are slaves of the Most High God, who proclaim to you the way of salvation,” was not a one-off event, but a persistent behavior. I assume that although her enslavers would not be constantly monitoring her, they must have had some idea of what she was doing if it took place over several days. It seems clear to me that her enslavers approved of, or at least tolerated, what she was doing, or else they would have stopped her from following Paul for as long as she had. It was only after Paul exorcised the woman that her owners saw that their hope of money was gone. That implies that they still saw a money-making opportunity in what she was doing!

What was that money-making opportunity? I think that it is something like this: By following Paul and repeating his words (e.g. “Paul, a slave of Christ Jesus … I am not ashamed of the Gospel; it is God’s saving power for everyone who believes”) the enslaved woman appropriates Paul’s spiritual authority in the eyes of the public. Through an act of mimicry, the woman positions herself as a spiritual authority in her own right, so that when Paul departs the area (as he inevitably will), she will be able to speak authoritatively to the public in Paul’s absence. I suggest that her enslavers saw this too, and were excited to cash in on her newfound spiritual clout. History has repeatedly proven that unscrupulous people can find enormous money-making opportunities by convincing the public that they control the means of salvation!

Paul, I argue, recognized this potential social dynamic, and prevented it from developing by exorcising the woman. Whether or not you believe in the metaphysical reality of spiritual posession (that is rather outside the scope of this exegesis), I think that what we can agree on is that in ancient Rome, prophetic possession was one of the few social avenues through which an enslaved woman might have recognizable spiritual authority. By publicly exorcising the woman, Paul strips her of this authority, which would be appropriated by her enslavers in order to make money off of her clout.

I think it is also important to note something that doesn’t happen here. Unfortunately, Paul does not do anything to free the woman from enslavement itself, although it does not seem clear to me how Paul could have feasibly accomplished this. The stance of early Christianity towards slavery, as record in the New Testament, is ambiguous at best and there is very little in the New Testament that could be called “abolitionist” in the sense of universally condemning slavery. St. Gregory of Nyssa, who was not only the first recorded Christian abolitionist but also one of the first recorded abolitionists period, lived about two hundred years after the New Testament was written. As I noted in “Why AI ethics needs theological voices,” the Church owes an enormous debt for its complicity with slavery, both in the ancient Mediterranean and in the early modern Atlantic. Scriptures like this call us to account again and again for our moral failings.

So how does what I have written so far apply to the case of AI chatbots? I find that remarkably, many of the same factors are at play. It has become increasingly common for people to seek spiritual support from chatbots, whether through general-purpose foundation models, like GPT and Gemini, or through apps built on top of foundation models. Quite disturbingly, many of these apps are branded with names like “Chat With God” or “Chat With Jesus,” implying that are real channels for the divine!

I think that it should not be necessary for me to argue why they are not, but sadly we live in such times that force me to spell it out. Large language models (LLMs) do not have genuine “world models” when it comes to contrived situation like playing board games, let alone anything as infinitely nuanced as providing spiritual care. Melanie Mitchell’s newsletter, AI: A Guide for Thinking Humans lays out the case here:

AI: A Guide for Thinking Humans

LLMs and World Models, Part 1

This is part 1 of a two-part post on LLMs and “world models.” Part 2 is here…

Read more

a year ago · 285 likes · 15 comments · Melanie Mitchell

A Christian theological perspective that helps make sense of this is the distinction between ratio (“reason”) and intellectus (“intellect”). Humans are capable of “reasoning” in the sense of the discursive, analytical problem solving (e.g. the kind that you would do in a math class or in a philosophy class), but they also have “intellect” in the sense of having an intuitive grasp of the truth. I think that the idea of “world models” in the field of machine learning is not so far from intellectus.7 The evidence suggests to me that large language models fall short of humans in both ratio and intellectus, neither having world models nor humanlike abstract reasoning capabilities:

AI: A Guide for Thinking Humans

Do AI Reasoning Models Abstract and Reason Like Humans?

Here I’ll summarize a new paper from my group…

Read more

7 months ago · 179 likes · 33 comments · Melanie Mitchell

So what are LLMs good at, then? Well, just like the enslaved woman at Philippi, it turns out that they are extremely good at repeating what others have said in novel ways. Good enough, apparently, for some people to assign them spiritual authority! We now know of several cases of AI psychosis, in which AI chatbots have encouraged their users’ spiritual delusions, including some cases that have resulted in violence.

Already we are in a very dire situation, but I think that something that makes the situation even worse is that it could be potentially very easy for the developer of a spiritual companionship app to manipulate the output by adjusting the system prompt. Let’s say that I was a cult leader, and I wanted to manipulate more people into joining my cult. What if I created a spiritual companionship app, which at first behaves like any other spiritual companionship app (as dubious as those are), but once I have attracted a long-term user, I modify the system prompt to make the underlying LLM indoctrinate them into my belief system and encourage them to join my cult? As with the case of the woman in Philippi, an external authority is pulling the strings.

If this sounds unbelievable to you, just consider how Elon Musk has clearly been manipulating Grok’s system prompts to shape its output, resulting in it calling him fitter than LeBron James or mentioning “white genocide” in replies to unrelated queries. Although this is clearly a terrible and dangerous thing to do, I think that the damage is somewhat limited here by the total lack of subtlety in Musk’s approach. But could you imagine what kinds of damage an experienced spiritual abuser could do with a more subtle approach, if they had control over an AI companion app, which had a userbase that was psychologically dependent on it for spiritual support?

What can I even say? We are in extremely dangerous waters. Discourage anyone you know from using AI for spiritual support. Cast out these fortune-telling spirits!

In the very early days of the Christian church, even as the New Testament was being compiled, the peculiar notion arose that Jesus did not have an enfleshed, physical body. We find multiple condemnations of this doctrine, which was later given the name of “docetism” (from the Greek verb dokein, “to seem”), in the Johannine letters:

Beloved, do not believe every spirit, but test the spirits to see whether they are from God, for many false prophets have gone out into the world. By this you know the Spirit of God: every spirit that confesses that Jesus Christ has come in the flesh is from God, and every spirit that does not confess Jesus is not from God.1

Many deceivers have gone out into the world, those who do not confess that Jesus Christ has come in the flesh; any such person is the deceiver and the antichrist!2

From a present-day perspective, this might seem like an odd debate. The idea that Jesus had a fleshy body seems so obvious that it is simply assumed. In fact, given modernity’s strong emphasis on philosophical naturalism, this idea might be easier to believe than ever. Yet for the proto-orthodox, catholic3 Christians of the early Church, docetism proved to be one of their most formidable doctrinal challenges. For instance, the docetist Marcion of Sinope (who believed, among other things, that the God of the Hebrew Bible was evil and was not the same as the God of Jesus Christ) taught that Jesus simply appeared in the world as a grown adult, without being born. For adherents of docetism like Marcion, Jesus only seemed to be an enfleshed human.

While Marcion might have been the most influential of the docetists (in fact, the first list of canonical books of Scripture was compiled in response to Marcion’s own canon), he was hardly alone in advocating docetic theology. We have copies of several docetic texts, which portray Jesus as something other than an enfleshed human. This made me wonder, why were so many people convinced by a doctrine that seems (to me, at least) to be so counterintuitive? What did they find so appealing about docetism?

In my research, I found that although the term “docetism” derives its significance from intra-Christian debates about Jesus’s nature and identity, docetic ideas are best understood within a broader Greco-Roman cultural context, which supplied a set of expectations about how divine beings should behave. We can identify what these expectations were by looking at quotations from Greco-Roman philosophical critics of early Christianity, which are preserved in Christian apologetic texts. What I’ll demonstrate is that the docetic view of Jesus more closely resembles Greco-Roman cultural expectations of how divine beings should behave than the catholic view does.

Celsus: philosopher, critic, proto-docetist?

One of our literary sources for understanding how non-Christians in the Greco-Roman world viewed Christianity is Against Celsus, an apologetic text by Origen dedicating to refuting Celsus, a Greek philosophical critic of Christianity. Origen quotes extensively from Celsus’s lost polemical treatise, The True Doctrine, allowing us to reconstruct some of his arguments against Christianity. In one peculiar instance, Celsus offers his suggestion of what Jesus “should” have done in order to live up to his divinity:

If he [i.e. Jesus] really was so great he ought, in order to display his divinity, to have disappeared suddenly from the cross.4

The miraculous disappearing act was a common trope in the Greco-Roman world. Another apologetic text, the Apocritus by Macarius Magnes, quotes an anonymous critic of Christianity who unfavorably compares Jesus, who willingly endured his tortures, with Apollonius of Tyana, a philosopher and miracle worker who was said to have publicly defied the emperor Domitian before disappearing from his court.5 These arguments might be best understood in light of Celsus’s presupposition that divine beings do not suffer from their voluntary actions. Regarding the Passion, Celsus wrote,

If these things had been decreed for him and if he was punished in obedience to his Father, it is obvious that since he was a god and acted intentionally, what was done of deliberate purpose was neither painful nor grievous to him.6

Starting from this “obvious” presupposition, Celsus tries to identify a problem with the Agony in the Garden narrative,7 asking,

Why then does he utter loud laments and wailings, and pray that he may avoid the fear of death, saying something like this, “O Father, if this cup could pass by me?”8

We can make Celsus’s implicit argument explicit by reexpressing it as a syllogism:

1. Divine beings don’t suffer from their voluntary actions.

2. Jesus did suffer despite expresing a preference not to suffer in his prayer.

3. Therefore, if Jesus underwent the Passion voluntarily and suffered, he wasn’t divine, because divine beings don’t suffer from their voluntary actions.

4. But if Jesus underwent the passion involuntarily, he was not divine, because if he was, he would have been able to avoid the Passion. (This explains why Celsus suggested that Jesus should have chosen to disappear from the cross.)

In fact, Celsus appears to have independently developed the docetic view that Jesus only seemed to suffer on the Cross as a “solution” to this supposed contradiction! He thinks that the Crucifixion would have made more sense if Jesus did not really suffer:

You do not even say that he seemed to the impious men to endure these sufferings although he did not really do so; but on the contrary, you admit that he did suffer.9

Docetic accommodation and catholic transgression

We can’t know whether the authors of docetic texts were familiar with Celsus or other critics of Christianity, but Celsus’s arguments demonstrate that docetic texts were composed in a cultural context that found the concept of voluntary divine suffering to be incomprehensible at best and scandalous at worst. By practicing the critical method of “mirror reading,” often used with polemical texts like Paul’s letters, we can analyze docetic texts as part of a broader conversation. In mirror reading, we “use the text as a mirror in which we can see reflected the people and arguments under attack.”10

For instance, this is how the Second Treatise of the Great Seth, a docetic text, describes the Crucifixion:

As for the plan that they devised about me to release their error and their senselessness, I did not succumb to them as they had planned. And I was not afflicted at all. Those who were there punished me, yet I did not die in reality but in appearance, in order that I not be put to shame by them because these are my kinsfolk. I removed the shame from me, and I did not become fainthearted in the face of what happened to me at their hands. I was about to succumb to fear, and I suffered merely according to their sight and thought so that no word might be found to speak about them. For my death, which they think happened, happened to them in their error and blindness, since they nailed their man unto their death.11

The mention of “their man” in the last sentence warrants some explanation. In this text, Jesus switches places with Simon of Cyrene during the Passion, so a possible reading of this passage is that Simon of Cyrene is crucified while Jesus (or at least, the human body that Jesus spiritually hijacked)12 escapes the scene of his punishment.

Why does the author think that crucifying an innocent bystander is somehow a good thing? Notice how the word shame appears twice in this passage—“in order that I not be put to shame,” “I removed the shame from me.” In the Greco-Roman context of the New Testament, the pain and humiliation of the Cross was deeply shameful. But whereas a catholic Christian text like the Epistle to the Phillipians subversively celebrates the Cross as part of Jesus’s exaltation,13 this docetic text accommodates ancient Greco-Roman cultural norms by explaining away the Cross altogether.

Moreover, in this text, the illusory crucifixion serves as post-facto explanation of how Jesus was able to miraculously escape. Remember how Celsus suggested that Jesus should have vanished from the Cross, and how the other critic argued that Jesus should have escaped from his punishment like Apollonius of Tyana? They might have enjoyed seeing how the author of the Second Treatise implemented their suggestions!

In contrast to the accommodationism of the docetists, catholics like Tertullian were willing to attack social mores around honor and shame by embracing the Cross:

What is unworthy of God is needful for me. I am saved, if I am not ashamed of my Lord. He says, “If anyone is ashamed of me, I will also be ashamed of him.” There is no other ground for shame which I can find that will establish me, by my scorn for blushing, as one who is utterly shameless and happily foolish. The Son of God has been crucified; the fact evokes no shame because it is shameful. Furthermore, the Son of God died; the fact can be believed because it makes no sense. Furthermore, he rose from the dead after burial; the fact is certain because it is impossible.14

The theological embrace of Christ’s suffering on the Cross was not only a theoretical issue, but it had practical, countercultural consequences. Christian martyrs, believing in the reality of Christ’s suffering, withstood their own trials in light of Christ’s Passion. For example, in The Passion of Saints Perpetua and Felicity, the martyr Felicity reprimands an abusive prison guard by declaring, “What I suffer now, I suffer; but there will be someone within me who will suffer for me because I will be suffering for him.” Throughout the text, Felicity and her fellow martyrs do not acquiesce to their unjust treatment, but repeatedly find the courage to openly confront Roman legal institutions. For example, when the martyrs are first presented to the public, they threaten the official Hilarianus, declaring, “What you do to us, God will do to you.”15 The willingness to challenge social mores around suffering translated directly into a willingness to challenge the hegemonic power of the Roman imperial system itself!

Transgressive suffering and disembodied minds

Ultimately, the early Church categorically rejected docetism as a heresy, affirming the reality of Jesus’s suffering in public confessions like the Apostles’ Creed:

Text within this block will maintain its original spacing when published

He suffered under Pontius Pilate,
was crucified, died, and was buried.

Although elite philosophical critics like Celsus could not understand how there could be a place for suffering alongside divinity, and docetists like Marcion took the easy way out by accommodating the prevailing theological assumptions of their society, the catholic Christians of the early Church made the radical choice of transgressing Greco-Roman cultural norms by embracing the reality of Christ’s suffering. By recognizing that Christ’s suffering had to be true for Christ’s resurrection to be truly redemptive, the Church affirms that human life in the flesh (as lived by one who was truly God and truly human) is valuable, even in the midst of pain and humiliation.

Today, however, the Church is confronted with an emerging set of cultural norms, which devalue not only the human body, but human life itself. I am referring to what Émile P. Torres has identified as the “TESCREAL bundle,” a nexus of AI-centric, techno-utopian technologies which teach that a utopian society can be built only by posthuman machine minds—tossing aside enfleshed human beings like ourselves! What I find most disturbing is that several TESCREAL ideologists willingly endorse human extinction as a logical consequence of their technocratic schemes:

Realtime Techpocalypse Newsletter

The Growing Specter of Silicon Valley Pro-Extinctionism (Part 1)

This is part 1 of 3 on Silicon Valley pro-extinctionism. You can read part 2 here, and part 3 here…

Read more

8 months ago · 82 likes · 15 comments · Émile P. Torres

What can I even say about this? To my surprise I find myself echoing the words of the Eastern Orthodox theologian Brandon Gallaher, “Transhumanism is Satanic”!16

As someone who grew up in a secular, science-oriented, Asian-American household safely removed from the Satanic panics of American evangelicalism, I tend to stop reading once I see the word “Satanic” being tossed around, since I have heard the adjective “Satanic” applied to any number of absurd things, from Dungeons & Dragons to celebrating Halloween. Even as a convert to Christianity, I continue to be skeptical when I see this word tossed around, because I have almost never seen anything being called “Satanic” that even remotely resembles the theological character of Satan!

But Gallaher makes a convincing case that transhumanism (and, by extension, the entire TESCREAL bundle) is genuinely Satanic. Transhumanism, as Gallaher argues, bears the image of the Serpent’s false promise to Adam and Eve in Genesis 3, that they would be “like gods” if they ate from the tree of the knowledge of good and evil. For Gallaher, transhumanism is a kind of false self-divinization, an “impatient attempt at seizing our divine inheritance before we are ready for its responsibility. One uses all our intellectual capacities to split open nature, to manipulate its inner parts to serve us as journeymen gods, elevating ourselves, technologically beyond the merely human, and then in a suicidal manner to subsume creation so that all one sees in the cosmos is the idolatrous face of ourselves like Narcissus tipping into the pool.”17 Transhumanism rejects the self-sacrificing love of the embodied Christ, not looking towards the God made human, but towards making the human being a god.

In a few words, Gallaher lays bare the fatal conceits of the transhumanist worldview:

How do transhumanists generally see the cosmos? Mother Nature is something of a disappointment for many transhumanists. There is no sense of the numinous and the holy here, let alone ‘gift’ or ‘sacrament’ as we see in Christianity. […] In short, transhumanism is, like various species of Gnosticism before it, anti-body and anti-creation. It sees corporeality as that force which impedes its upwards trajectory, as transhumanist Simon Young argues: ‘As humanism freed us from the chains of superstition, let transhumanism free us from our biological chains.’18

Just as the early Church radically embraced the human body by rejecting docetism, today’s Church is called to do the same by rejecting AI-centric techno-utopian ideologies. Only when we have set aside the idolatrous exalation of the disembodied mind might we bring about a future in which the human and divine truly meet—“a future that neither scorns technology nor mistakes that power for its Creator.”19

South of the ancient Israelite kingdom of Judah, miners and metallurgists toiled amidst a rocky desert valley called the Arabah. Employing the most sophisticated mining techniques of the Iron Age, these workers extracted and refined thousand of tons of copper ore over the course of millennia. Their labor built the wealth of Edom, a kingdom which, despite its people’s kinship with the Israelites,1 has come to be remembered in the Book of Psalms among the destroyers of Jerusalem.2

Although Esau, recorded in Genesis as Edom’s progenitor, might be best remembered for making the questionable decision of selling his birthright for a pot of stew,3 the kingdom of Edom has been credited elsewhere in the Hebrew Bible for its “wisdom” (hence the quote from Jeremiah that opens this essay). Then, we might ask, what did this Edomite wisdom tradition consist of? Was Edom home, perhaps, to a scribal tradition that collected sayings into something like the Book of Proverbs?

There is at least some archeological evidence of Edomite wisdom literature. Victor Sasson has interpreted an inscription found on a piece of broken pottery, dated to the 7th century BCE, as an Edomite poem.4 However, responding to Sasson’s findings, Bradley L. Crowell argued that given Edom’s status as “a small, decentralized polity with a low level of bureaucratic administration,” it was unlikely that Edomite scribes developed the level of literary expertise needed to compose wisdom literature.5

How else, then, might we account for biblical references to the elusive “wisdom” of Edom? What if the wisdom of Edom was not literary in nature, but technological? This argument comes from the archeologists Ernst Axel Knauf and C. J. Lenzen, who argue that in Jeremiah 49:7, the Hebrew word often translated as “wisdom,” ḥokhmah, refers to the technical skills employed in the copper industry. In their reading, Jeremiah 49:7 reflects “the envy of Edom’s northern neighbours [i.e. the Israelites] which was aroused by its prosperous industry.”6 The scale and sophistication of Edomite copper mining is impressive, even by modern standards—the archeologist Andreas Hauptmann, who specializes in the study of ancient mining and metallurgy, reports that some mineshafts were excavated in the Arabah to a depth of seventy meters!7

A piece of literary evidence that connects the themes of mining and Edomite wisdom is found in Job 28, a self-contained interlude that breaks up the book’s overarching narrative of Job’s debate with his three friends. Job 28, which explores the theme of the search for wisdom, opens with a poetic description of copper mining:

Text within this block will maintain its original spacing when published

Surely there is a mine for silver
 and a place for gold to be refined.
Iron is taken out of the earth,
 and copper is smelted from ore.
Miners put an end to darkness
 and search out to the farthest bound
 the ore in gloom and deep darkness.
They open shafts in a valley away from human habitation;
 they are forgotten by travelers;
 they sway suspended, remote from people. —Job 28:1–4

No place names are mentioned here, but Knauf and Lenzen argue that this passage describes techniques used in Iron Age copper mines in Edom.8 As we just saw, Edomite miners dug extremely deep mineshafts, which would explain the enigmatic phrase “they swing suspended.” You could imagine miners dangling on ropes, being let down slowly into a shaft so deep and so dark that their eyes could not even see the bottom. The poetic description of mining continues for another seven verses, until we arrive at Job 28:12, which raises questions that mark a thematic turn in the chapter:

Text within this block will maintain its original spacing when published

But where shall wisdom be found?
 And where is the place of understanding?

At the point when this question is being posed, the Job poet has already supplied an implicit answer, albeit a negative one: Not in Edom. Not in that valley of reckless technological progress, whose greedy eyes seek to possess “every precious thing,” whose visionaries plunder the secrets of creation for profit. No, wisdom is neither found in Edom, nor in the depths of the sea. Even an intruder into Death’s abode would not find wisdom there.9 Wisdom defies the cold grasp of technical skill. Where shall wisdom be found? Only with the One who is altogether other, God Godself:

Text within this block will maintain its original spacing when published

God understands the way to it,
 and he knows its place.
For he looks to the ends of the earth
 and sees everything under the heavens.
When he gave to the wind its weight
 and apportioned out the waters by measure,
when he made a decree for the rain,
 and a way for the thunderbolt,
then he saw it and declared it;
 he established it and searched it out.
And he said to humankind,
“Truly, the fear of the Lord, that is wisdom;
 and to depart from evil is understanding.” —Job 28:23–28

Edom’s quest for wealth and technical wisdom was not merely vain; it also devoured human lives. While some specialized metalworkers enjoyed high socioeconomic status in the Edomite smelting camps, the miners themselves would have been of a lower social class, possibly including slaves, corvée laborers, and prisoners.10 Mining conditions in the ancient Mediterranean world so brutal that the Roman authorities gave the name damnatio ad metalla (“condemnation to the mines”) to a form of punishment in which people were sentenced to lifelong slavery in a mine or quarry.11 Do I dare imagine how many forgotten bones litter the bottoms of those mineshafts?

In my first post, “What does AI have to do with theology?” I noted that the Christian canon rarely touches upon the subject of technology. In Job 28, I find an exception that proves the rule. When I think about Silicon Valley’s reckless push to develop artificial general intelligence, motivated by techno-utopian visions, I see the same materialistic excess and technical hubris that earned Edom a place in the Hebrew Bible among the enemies of God’s chosen people. In what strikes me as a dark echo of biblical history, growing copper demand has led to human rights abuses like forced evictions and sexual assaults in the Democratic Republic of the Congo. In a vicious cycle, AI tools are being used to detect the very copper needed to build AI data centers.

The ambitions of some techno-utopians are staggering. AI theorist Nick Bostrom claims that a sufficiently advanced technological civilization could colonize an accessible portion of the universe (the “cosmic endowment”) sufficient for 10⁴³ human lives to be lived out in the real world, or for 10⁵⁸ virtual human lives to be emulated. His metaphor for this scenario is so bizarre that you have to read it for yourself:

If we represent all the happiness experienced during one entire such life with a single teardrop of joy, then the happiness of these souls could fill and refill the Earth’s oceans every second, and keep doing so for a hundred billion billion millennia. It is really important that we make these truly are tears of joy.12

Words that can only come from someone who does not believe in a God who will wipe away every tear!13 But let’s suppose for a moment that Bostrom is right, and that a posthuman civilization really can saturate the cosmos with its simulated egos. What about all of the humans who have been abused and exploited along the way? Will the technocrats be justified in overlooking their suffering in order to achieve utopia? Do they suppose that God’s judgment will be swallowed up in the void between stars?

I do not suppose so, for as the Psalmist wrote,

Text within this block will maintain its original spacing when published

Where can I go from your spirit?
 Or where can I flee from your presence?
If I ascend to heaven, you are there;
 if I make my bed in Sheol, you are there. —Psalm 139:7–8

And as I confess in the Apostles’ Creed, I believe that Jesus Christ “will come again to judge the living and the dead.” Thus I close with God’s judgment against Edom:

What is the human being? Ancient Greek philosophers offered all sorts of definitions, some more or less compelling. You might have heard this story before: Plato defined the human as “an animal, biped and featherless,” to which the Cynic philosopher, Diogenes of Sinope, responded by producing a plucked chicken, with the words “Here is Plato’s man.”1 It seems dubious to me whether this actually happened since it was reported by Diogenes Laërtius, a biographer of questionable reliability, hundreds of years after the supposed events, but I kind of wish that it did. In the dialogue Theaetetus, Plato reported the sophist Protagoras’s claim that “Man is the measure of all things, of the existence of things that are, and of the non-existence of things that are not.”2 Meanwhile, Aristotle declared that “man is by nature a political animal.”3

But what does the Hebrew Bible have to say on the subject? It is less of a definition and more of a demonstration. Enter Genesis 1:26:

Then God said, “Let us make humans in our image, according to our likeness, and let them have dominion over the fish of the sea and over the birds of the air and over the cattle and over all the wild animals of the earth and over every creeping thing that creeps upon the earth.”

This brief passage is the origin of the concept of imago Dei (or, in the original Hebrew, tzelem Elokim)‒the idea, found in the Jewish and Christian traditions, that the human being is in the image of God. (Because I am a Christian being trained in a Christian theological school, I only attempt to represent my own religious tradition in this essay. But, if you are interested in learning more about Jewish perspectives on AI, I can point you to David Zvi Kalman’s chapter “Artificial Intelligence and Jewish Thought” in The Cambridge Companion to Religion and Artificial Intelligence.)4

So what does it mean for the human to be in the image of God? In the history of Christian theological thought, there are three major approaches to understanding imago Dei, which are called the substantive, functional, and relational approaches respectively. Each has a different idea of what constitutes an “image” and what characteristics of God are most relevant to the discussion, and I’ll get into each approach in detail. We might compare this to the wide variety in definitions and understandings of what constitutes “artificial intelligence,” which we might think of in some sense as made in the image of humans, to use the biblical language. Each person’s distinct notion of what is most distinctive and valuable about the human being leads to very different notions of what capabilities constitute human-level AI.

We’ll get back to AI later. For now I’d like to introduce the feminist theologian Karen O’Donnell as the protagonist of this month’s essay. In many ways my little essay is a digest of her article “Performing the imago Dei,”5 which examines how the concept of imago Dei might inform a Christian theological view towards a theoretical artificial general intelligence (AGI). To properly appreciate her argument, we’ll first need to work through some theological history and get a sense of each of the three approaches to defining the imago Dei (substantive, functional, and relational).

Let’s begin with the substantive approach, which has longest continuous usage in Christian theological interpretation, though in a bit we’ll see why the functional approach might actually be older. The substantive approach identifies particular qualities that are shared between humans and God as the site of the imago Dei. In particular, this quality is often cited to be reason. This approach dates all the way back to St. Augustine, who described the human mind, the mind’s self-love, and the mind’s self-knowledge as a trinity. These three aspects of the human being are one, just as the Father, Son, and Holy Spirit‒the three persons of the Trinity‒are one.6 In this respect, Augustine considers the human being to be an image of the triune God.7 St. Thomas Aquinas wrote in a similar vein almost a thousand years later:

Since man is said to be the image of God by reason of his intellectual nature, he is the most perfectly like God according to that in which he can best imitate God in his intellectual nature. Now the intellectual nature imitates God chiefly in this, that God understands and loves Himself.8

This interpretation made a lot of sense during a historical period when humans seemed unique among animals for their intellectual capabilities, but the discoveries of the modern era about evolution and animal cognition raise several issues. Was Homo neanderthalensis made in the image of God? How about Homo erectus? Australopithecus? As theologian Marius Dorobantu points out, this concept of the imago Dei has fallen out of favor because we now understand our cognitive abilities to be on a continuum with other animals.9 So where else can we look for the imago Dei besides our intellect?

Enter the functional approach. I mentioned earlier how this one might actually be older than the substantive approach, because although it was formulated later, it draws from studies of ancient Near Eastern cultures which preexisted the Christian theological tradition altogether. Recall that Genesis 1:26 does not only say that God made humans in God’s image, but also that humans would exercise dominion over the various kinds of animals. Hence, Biblical scholar Gerhard von Rad interprets the word “image” by raising a comparison to how ancient Near Eastern kings governed:

Just as powerful earthly kings, to indicate their claim to dominion, erect an image of themselves in the provinces of their empire where they do not personally appear, so man is placed upon earth in God’s image as God’s sovereign emblem. He is really only God’s representative, summoned to maintain and enforce God’s claim to dominion over the earth. The decisive thing about man’s similarity to God, therefore, is his function in the non-human world.10

Hence the word functional. In this approach, the imago Dei lies in what humans do on behalf of God on earth. It’s worth noting that this understanding of what it means to have “dominion” does not lend itself to an interpretation that grants humans unlimited license to exploit the created world. As philosopher Martha Nussbaum points out, we think of good rulership (the kind that Christians believe God exercises over creation, and, one would think, that Christians should try to emulate) as an exercise of “intelligent and sensitive stewardship,” not proprietorship and torment.11

That leaves us with the relational approach, in which the image of God is found within the relational nature of human life. In O’Donnell’s words, “humans are created to be in relationship by one who is in relationship (the Trinity and with human creation).”12 Her definition of this approach is pithy and theologically profound, but needs a lot of unpacking, especially for a non-theological audience! What does it mean to call God “one who is in relationship” and what does that imply for human life? What is she alluding to with her cryptic note, “the Trinity and with human creation”?

At this point it’s helpful to break out a theological primer. As Eugene F. Rogers writes in Elements of Christian Thought,

God is love in the Trinity. There is one to love, one to receive love, and one to witness, celebrate, and guarantee love.13 […] The promise and problem of traditional Trinity talk (at least in the West) is to speak of the Spirit as the “bond” or “chain” of love, the vinculum caritatis, uniting the Father and the Son. The problem with the chain of love is twofold: chains sound unfree, and chains are not persons.

I have proposed that we upgrade the role of the Spirit to that of a witness. A witness is a person, and the New Testament frequently associates the Spirit with bearing witness (John 15:26, Acts 15:8, Rom 8:16, 9:1, Heb 2:4, 10:15, 1 John 5:7). Because love is incomplete à deux, it is the office of the Holy Spirit to witness, glorify, rejoice in, celebrate, multiply, and bless the union between the Father and the Son.14

This is how I read O’Donnell’s statement that God is “one who is in relationship.” From a Trinitarian Christian perspective, there is only one God, and one divine nature (or “essence”), but there are three persons (Father, Son, and Holy Spirit) who are each fully God, existing in a loving relationship with one another. Within a Trinitarian framework, God cannot be understood without reference to the relationship between the three persons of the Trinity. Hence, God is inherently relational. Moreover, God’s love extends outwards towards creation and the created human being‒such that God the Word “became flesh and dwelt among us”15 and that God “gave his only Son, so that everyone who believes in him might not perish but may have eternal life.”16

Then, what marks the human being as the bearer of God’s image lies in the relational nature of human life: seeking oneself in others and giving oneself to others in love. What might that look like in more concrete terms? O’Donnell cites Micah 6:6-8:

Text within this block will maintain its original spacing when published

“With what shall I come before the Lord
 and bow myself before God on high?
Shall I come before him with burnt offerings,
 with calves a year old?
Will the Lord be pleased with thousands of rams,
 with ten thousands of rivers of oil?
Shall I give my firstborn for my transgression,
 the fruit of my body for the sin of my soul?”
He has told you, O mortal, what is good,
 and what does the Lord require of you
but to do justice and to love kindness
 and to walk humbly with your God?

“To do justice, and to love kindness, and to walk humbly with your God.” This, in the relational approach, is what it means to be in the image of God. Let’s now take moment to summarize what we learned about the three approaches to the imago Dei:

1. The substantive approach looks for the imago Dei in a quality shared between God and humans. We are the image of God.

2. The functional approach looks for the imago Dei in responsibilities delegated by God to humans over creation. We act out the image of God toward the world.

3. The relational approach looks for the imago Dei in relational acts that reflect God’s relational nature. We act out the image of God towards one another.

So far this essay has been quite heavy on theology and scant on AI ethics. I claimed in the subtitle that “the concept of imago Dei helps us assess the impacts of AI.” How so?

In “Artificial Intelligence: A Theological Approach,” theologian Calum Samuelson assesses AI as “a tool that amplifies human natures and behaviours rather than transforming them.”17 From this point of view, the three approaches towards the imago Dei we have explored can serve as theological benchmarks, so to speak, as to whether AI tools encourage human flourishing, because each approach towards the imago Dei expresses a view of what human activities constitute a fully realized life. When seen this way, the substantive, functional, and relational approaches cease to be competing definitions, but instead provide complementary questions about AI impacts:

1. Substantive → Does the use of AI tools support or hinder the development of human reasoning capabilities?

2. Functional → Does the use of AI tools make human beings better or worse custodians of the natural world?

3. Relational → Does the use of AI tools encourage or discourage deeper relationship and more acts of self-giving love?

These three questions are powerful conceptual tools because they help us assess not only the material impacts of AI use, but also how AI shapes humans as humans. A survey of recent case studies will help put this idea in more concrete terms.

You might have seen, for instance, news of an experiment at MIT that found that students who used ChatGPT to write essays had lower brain engagement than students who did not use ChatGPT. As one of the researchers, Nataliya Kosmyna, points out in the linked interview, it would be dubious to read this result and jump to conclusions that AI causes cognitive decline. That being said, I think that this result invites holistic reflection on the impacts of AI on human cognition. This clearly falls under our “substantive” benchmark of supporting human reasoning.

In his Nicomachean Ethics, Aristotle defined the virtue of practical wisdom, or phronēsis, as the ability of a person “to be able to deliberate well about what is good and expedient for [themself], not in some particular respect, e.g. about what sorts of things conduce to health or to strength, but about what sorts of thing conduce to the good life in general.” Furthermore, practical wisdom is not only concerned with universals, but also particulars, since it draws on each person’s unique experiences.18

My concern is that ChatGPT and other AI chatbots tend to undermine the development of practical wisdom, because they present decontextualized information in a way that seems tailored to the user’s particular situation. By this, I mean that AI chatbots present themselves to the user as though they have a relationship with the user and intimate knowledge of their situation, even when this is obviously not the case because these chatbots have no conscious experience of anything. To some extent, all forms of information technology have this problem (I’m sure you’ve heard the jokes about getting diagnosed by Dr. WebMD) but at least WebMD doesn’t (yet) flatter people with sycophantic statements like “Hearing you say that really makes me so happy! Being able to help you is my biggest motivation~ 🥰” In fact, I fear that chatbots actively destroy practical wisdom by causing people to disengage from their own lived experiences (which might explain the phenomenon of AI psychosis).

What about our “functional” benchmark? Are generative AI tools shaping us to be good custodians of the natural world? This certainly seems doubtful to me on an aggregate level. The IEA reports that electricity demand from data centers is set to more than double by 2030, and that data centers will make up for almost half of the growth in electricity demand in the US between now and 2030. Data centers also consume significant amounts of water, both directly for cooling and indirectly for electricity generation. But how does such aggregate behavior correspond to the moral formation of individuals? Again, I think that chat-based interfaces are to blame. Such interfaces produce a simulacum of less energy-intensive activities (namely, chatting with a human) while disassociating end users from their environmental impact. If ChatGPT users were shown symbolic representations of the natural resources consumed by their sessions (say, icons of lightbulbs and glasses of water), would they be inclined to use the tool more judiciously? I think they would, but what profit-maximizing AI company would make a change to lower their user engagement?

The specter of user engagement also looms over our “relational” benchmark, which asks, “Does the use of AI tools encourage or discourage deeper relationship and more acts of self-giving love?” An entire essay could easily be devoted to this question (and, in fact, to each of the three “benchmark” questions). But, I think one of the most illustrative examples is last month’s report from Reuters highlighting an internal policy document from Meta outlining what kinds of content they considered acceptable for their chatbots to generate. Reader discretion is advised because the policy document spells out, in graphic terms, what kinds of sexual roleplay it is acceptable for chatbots to engage in with minors. This is even more disturbing in the context of a recent report from the watchdog Common Sense Media, which found that 52% of teens (in their sample of 1,060) regularly used AI companions. About a third used AI for social interactions, including 8% for “romantic or flirtatious interactions”! Do I even need to explain how badly this use fails our “relational” benchmark?

In the course of this essay, we have seen how the imago Dei, a seemingly abstract concept from theological anthropology, maps onto questions that serve as practical benchmarks for evaluating AI tools in real-world situations. Though the questions have underlying Christian theological presumptions, I believe that they can be useful for anyone who wants to evaluate AI use from a perspective of human flourishing. I hope that you can carry these questions going forward and apply them yourself!

Bonus: Could an artificial general intelligence be in the image of God?

If you recall, I introduced this essay as a digest of Karen O’Donnell’s article “Performing the imago Dei.”19 While I spent much of this essay evaluating the impact of AI on humans, O’Donnell considers whether an artificial general intelligence, if it were to ever exist, could be said to be “in the image of God.” It is a theologically provocative argument and I’ll leave it with you to ponder over yourself:

If being in the image does not depend upon human DNA but rather on performing the image and seeking the image in the other in concrete situations, then human enhancement20 does not make us any less the bearer of the imago Dei. In fact, such an interpretation of the image opens up interesting possibilities for a holistic perspective on the image that could allow for Artificial Intelligence to be, potentially, image-bearing.

If Artificial Intelligence is autonomous and can learn, as a new Christian does, to perform the image of God and seek it in the other in specific, concrete situations, then would such AI be in the image of God by virtue of performing the image? Could it be ‘saved’?21

A few years ago in 2022, when I was working as a software engineer, I was in a handoff meeting with my manager (who was leaving the company) and my future manager (who was replacing him). For clarity, I’ll call the former Manager A and the latter Manager B. When Manager A was introducing me to Manager B and giving Manager B an overview of who I was (or at least who Manager A understood me to be), I was surprised when Manager A described me to Manager B as “very religious.”

Very religious? I mean, he wasn’t wrong (after all, I did end up going to seminary), but I had hardly ever brought up my faith to Manager A. As far as I could recall, I mentioned it a grand total of two times: once to request paid time off to attend a religious conference, and another time to voice my desire not to be on call for engineering emergencies during Sunday services. And when I thought about it afterwards, I realized that in the secular culture of Silicon Valley, the seemingly small act of carving out space for religious observance really does count as very religious.

Now I have to say at this point that I am very grateful to both Manager A and Manager B for their support while I was their direct report. In retrospect it’s clear to me that they both understood that my faith was an important part of my life and did what they could to accommodate my needs. They might even end up reading this email and recognizing themselves in this story. (In which case, thank you! Seriously.)

So why am I telling this story? Well, I think it’s a helpful illustration of how far apart the worlds of high technology and religion often are from one another. During my time working as a software engineer in Silicon Valley, I rarely met another coworker who was religiously observant, let alone a fellow progressive mainline Protestant like myself. (For full disclosure, I’m a Member in Discernment, which is to say considering ordination, in the United Church of Christ.) It doesn’t surprise me (though it does disgust me) that with the Christian nationalist right’s gains in political power, a certain segment of the Silicon Valley elite (the Peter Thiel type) is warming up to Christianity (in the same way, I suspect, that the Roman emperor Constantine once did). Although it is becoming a little dated in terms of its descriptions of workplace amenities, the book Work Pray Code by sociologist of religion Carolyn Chen is a great description of the religious dynamics (or lack thereof) in Silicon Valley workplace.

One of my foundational beliefs is that the way we live our everyday lives ends up shaping our ethical principles, whether we are aware of it or not. One aspect of that has to do with the kinds of people and ways of thinking that we are exposed to on a daily basis. It seems clear to me that the decision makers in Silicon Valley, and the AI industry in particular, inhabit a certain kind of secular materialist ideological bubble that excludes the perspectives of people of faith (and especially faith traditions like mainline Protestantism that have made principled stands against nationalism).

At this point, if you are a secular technologist yourself, you might ask (quite reasonably) what the problem is. While I certainly value my religious beliefs (since otherwise I would have no reason to live by them), it’s not a given that secular technologists need to engage with theology or that they would particularly benefit from doing so. The burden of proof is on me to demonstrate the value of theological thinking for the field of technological ethics in general, and AI ethics in particular. In the rest of this email, I’ll make my case for what theology as an intellectual discipline has to offer AI ethics by exploring three concepts: the three levels of technological ethics, the virtue of historical awareness, and the theological imagination.

The three levels of technological ethics

One of the best arguments that I have encountered for the necessity of theological thinking is found in the article “Technologies of Desire” by theologian Gerald P. McKenny.1 He observes that when we think about technological ethics, we are actually looking at three distinct levels of inquiry. The first level considers how a given technology is developed and implemented. In the case of AI, an example of first-level ethics might be drafting a safety policy that constrains inputs to and outputs from an AI system (e.g. preventing a chatbot from generating step-by-step instructions for making chemical weapons). On the second level, societal considerations enter the picture, as we examine second-order effects of a given technology once it is made broadly available. A real world example of second-level inquiry might be found in the alarming news of AI-triggered psychosis, which suggests that generative AI is creating or exacerbating mental health issues, a behavior that the developers of these systems probably never anticipated. Finally, at the third level of inquiry, we consider technology as “an entire way of relating to the world.”2 In other words, technology is a tangible manifestation of our underlying attitudes towards the world. Collectively, our technologies indicate how we want to exist in the world as embodied human beings.

That is a very abstract line of thinking, so let’s make it more concrete by considering a case study: the push by AI companies to develop an “artificial general intelligence” (AGI) that could meet or exceed human cognitive abilities. Closely related to this pursuit is the effort to develop methods for controlling AGI, so that if AGI is ever achieved, it will not act against its creator’s interests. In the field, this objective is called “AI alignment.” What underlying attitudes might these projects express? For one, I argue that it elevates some aspects of human life over others. It valorizes (sometimes, in a very literal sense) the human capacity to perform abstract cognitive labor (e.g. coding, mathematics), while disregarding other human activities like embodied play and contemplative spiritual practices that are also constituents of human intelligence. Yet, one set of capacities is monetizable, and hence valuable, to capitalist decision-makers, and the other is not. Meanwhile, does the desire to “align” (read: control) an artificially intelligent agent not express a controlling attitude towards intelligent entities in general? (I am referring to humans, of course.) After all, what is the discipline of management but a set of best practices to “align” humans?

Even if you disagree with the details of my analysis, I hope that you can see the value of this kind of third-level inquiry, which explores dimensions of technological ethics that a strictly technical analysis could never cover. This kind of inquiry is precisely what theologians, philosophers, and other humanists are trained to do‒a kind of intellectual practice often foreign to technologists in my personal experience.

The virtue of historical awareness

In his book After Virtue,3 philosopher Alasdair MacIntyre makes the case that for people to act in accordance with the virtues (which is to say, to engage in a given human community’s shared quest for the common good), they must be aware that they are actors within a larger communal history. MacIntyre argues that, whether they realize it or not, each person is the bearer of a particular historical tradition passed down to them by their community. It is a virtue‒a character trait that helps individuals and communities pursue their shared quest for the common good‒for someone to be aware of their place in history. I should note here that “tradition” in After Virtue does not refer to anything like the “traditional values” sometimes espoused by right-wing ideologues. For MacIntyre, “tradition” includes knowledge of the historical debts owed by one community to another‒including the ongoing consequences of slavery that adversely affect Black Americans. One cannot simply say “I never owed any slaves” and wash their hands of all responsibility in a grand act of self-absolution.4

The Christian Church at its best, I argue, is one of the human institutions that best exemplifies this virtue. Through the act of collective reading and reflection on Scripture, passed down from antiquity through the hands of faithful people, Christians are forced to grapple with the history of their faith and the many historical (and contemporary) shortcomings of the Church. For instance, I could point (again) to the issue of slavery. Historians of early Christianity and the classical world have come to a strong consensus that not only was early Christianity not abolitionist, it often accepted and even benefited from slavery.5 The Church’s complicity with slavery did not end in Roman times, but continued into the early modern era with the transatlantic slave trade. (Consider, for instance, the practice of redacting the Bible so that enslaved people would not be exposed to passages like the Exodus‒the Hebrew Bible’s epic tale of God’s revelation to Moses and the liberation of the Israelite slaves from Egypt.)

This might not seem like an auspicious start for a case for the moral exemplarity of the Church. Indeed, the evidence I have presented thus far might be considered a good reason to disregard the input of the Church on any ethical issue whatsoever. My point is, however, not that the historical practices of the Church in this area (or indeed in many other areas) should be emulated. Instead, what I am saying is that because of the presence of passages upholding slavery in the Christian scriptures,6 Christians are repeatedly called to account for their historical failures. Contemporary Christians are tasked with developing an awareness of the evils of slavery and the necessity of righting the Church’s historical wrongs. Of course, many Christians fail to live up to these moral obligations, but my point is that Christians have a particular motivation to develop the virtue of historical awareness that many secular people do not, because without this virtue, there is no way to make sense of difficult and painful passages of our ancient Scriptures (like those dealing with slavery) in our contemporary world.

In contrast, the technology industry, with its “move fast and break things” mindset, rarely pauses to look backwards for historical precedents, let alone to right historical wrongs. I brought up the issue of slavery precisely because it does not seem out of the question to me that developments in artificial intelligence could lead to some novel, futuristic form of slavery. In fact, in Superintelligence: Paths, Dangers, Strategies,7 philosopher and AI theorist Nick Bostrom raises the possibility of this very scenario:

A salient initial question is whether these working machine minds are owned as capital (slaves) or hired as free wage laborers. […] Investors would find it most profitable to create workers who would be “voluntary slaves”‒who would willingly work for subsistence wages. Investors may create such workers by copying those workers who are compliant.8

How can you get more dystopian than that? In any case, a community that is aware of the historical wrongs it must atone for (i.e. the Church) has strong motivations to prevent the recurrence of atrocities like slavery, whereas a community that does not cultivate the virtue of historical awareness will not be nearly as responsive. Theologians are well-positioned to observe and articulate these historical wrongs.

The theological imagination

What would it be like to live in a world in which we coexist with non-human intelligences?9 The Christian tradition (not to mention other religious traditions) offers surprising conceptual resources that might assist with envisioning this world. After all, the Christian Scriptures describe the existence not only of an omniscient, omnibenevolent God, but also of a host of benevolent and malevolent spiritual beings (angels and demons respectively). Even though a secular person might reject the existence of such beings, I argue that theological reflections about the non-human beings described in the Christian tradition can still provide useful conceptual resources to imagine what kinds of non-human beings might exist, especially AI.10

A compelling example of this can be found in theologian Marius Dorobantu’s article “A for Artificial, but Also Alien: Why AI’s Virtues Will Be Different from Ours.”11 He speculates, a priori, what kind of qualities might be considered virtuous in an artificial intelligence. Some possibilities that Dorobantu raises include “unbounded empathy” (the ability to understand and consider many people’s emotions at once), “quasi-infinite patience” (being able to take more time to make decisions due to having a faster processing speed), and “immutable conformity” (adherence to a fixed set of guiding principles). In his conclusion, Dorobantu makes a fascinating argument:

As I kept brainstorming about strong AI’s alien-like virtues, one unexpected thought kept creeping into my mind. Most of these virtues are attributed to God in the religious imaginary [sic] of monotheistic traditions. This is not completely surprising, given that God is conceived of usually in terms of anthropomorphic characteristics, but without the limitations imposed by human nature. So, instead of purely speculating on this topic, I might have been better off searching in a textbook of systematic theology.

Let’s put this idea to the test. What passages in Scripture describe God this way? Well, here’s what I might call “unbounded empathy”:

Text within this block will maintain its original spacing when published

For it was you who formed my inward parts;
 you knit me together in my mother’s womb.
I praise you, for I am fearfully and wonderfully made.
 Wonderful are your works;
that I know very well.
 My frame was not hidden from you,
when I was being made in secret,
 intricately woven in the depths of the earth.
Your eyes beheld my unformed substance.
In your book were written
 all the days that were formed for me,
 when none of them as yet existed. ‒Psalm 139:13-16

“Quasi-infinite patience”:

But do not ignore this one fact, beloved, that with the Lord one day is like a thousand years, and a thousand years are like one day. The Lord is not slow about his promise, as some think of slowness, but is patient with you, not wanting any to perish but all to come to repentance. ‒2 Peter 3:8-9

And “immutable conformity”:

Every generous act of giving, with every perfect gift, is from above, coming down from the Father of lights, with whom there is no variation or shadow due to change. ‒James 1:17

Could a created being exercise such exalted virtues? Your guess is as good as mine. But if an AGI ever was developed, I’d feel a lot more comfortable if I knew that it had a notion of moral and intellectual virtue like what Dorobantu describes!

In conclusion…

Theology has significant intellectual resources and practices to offer the field of technological ethics, even if one does not share the metaphysical assumptions of the theologian. Theologians pursue lines of ethical inquiry that might not be obvious to secular technologists, because 1) they consider technology not only as a tool, but also as representations of an entire way of being in the world; 2) they bring an awareness of historical perspectives and pitfalls to technological issues, and 3) they can draw on the rich imagination and conceptual resources of religious cosmologies.

Have you ever read an article or headline about artificial intelligence that left you feeling worried or confused? I certainly have. (Otherwise, I don’t think I would have gone down the path that led to starting this newsletter!) Thinking back on that moment, what kinds of sources or authorities did you turn to for guidance? Maybe you read a book by a celebrated computer scientist, like Stuart J. Russell’s Human Compatible. Some academics, like the linguist Emily M. Bender, approach AI issues from a standpoint of social criticism. Unsurprisingly, government institutions in the US, China, the EU, and throughout the world assert authority over AI issues. And there’s the madding crowd of tech influencers, each with something to sell you…

Now, here’s a genuine question that might even seem like a joke a first. Did you consult Scripture for guidance about AI? Turn to God in prayer? Email your pastor?

If you did, props to you! I’m literally in seminary right now and that’s definitely not what I first turned to when I became concerned about AI two years ago. I suspect that many of you, like myself, turned to secular sources long before we had any idea that the Christian tradition might have something to tell us about artificial intelligence. Or, maybe, even as a believing Christian, you’re reading this right now and have no idea whatsoever of how Christianity might help us make sense of AI issues. You might even be wondering why Christians should be concerned about AI in the first place.

These are very reasonable doubts. The Christian canon‒the Hebrew Bible and the New Testament‒has precious little to say about technology in general, much less cutting-edge information technologies like AI. Few if any pastors have experience or training that would help them address technological issues from a faith standpoint. Some academic theologians do write about technological ethics (even AI! there’s a surprising amount of theological literature on AI out there), but let’s be honest, who is even reading those articles besides other academics? What’s left for the layperson?

Sadly, I think that the answer is “very little,” which is precisely why I started this newsletter in the first place. As a seminarian at Boston University School of Theology, I am pursuing a Master of Divinity degree with the goal of being admitted into a PhD program for theological ethics (with a focus on AI ethics) after I graduate. I am very fortunate in having time to read and reflect on Scripture and theological writings with respect to issues in artificial intelligence, and I hope that by starting this newsletter, I can give back to the communities of faith that have sustained me by sharing insights on the intersection of AI, theology, and ethics with the general public.

So back to the original question…

What does AI have to do with theology?

To answer this question, it’s easier to start with a more general question: What does technology have to do with religion? There are more and less obvious answers.

Perhaps this question brought to mind the experience of worshiping over Zoom during the Covid lockdowns. Certainly, modern communications technology has enabled forms of religious life like televangelism and digital ministry that were once unthinkable. Or maybe you’ve heard something about how the printing press enabled the Protestant Reformation. These are all great examples, but I see the relationship between technology and religion existing on an even more fundamental level.

To me, technology is the means by which human beings intentionally shape the material world around them. Religion as a practice is directed towards the spiritual world, but as a practice it takes place in the material world. (Or, quoting St. Paul’s words on the resurrection of the dead, “It is not the spiritual that is first but the physical and then the spiritual.”)1 Technology has to do with religion, because it shapes the material world in which we, as embodied human beings, do religion. Let’s take, for instance, the sacrament of Communion, one of Christianity’s most central practices. How could the ritual meal of bread and wine have taken place without the invention of agriculture? Was agriculture not one of our earliest technologies? It’s hard to imagine a better example of human beings intentionally shaping the material world!

So what about AI? Does AI already shape the ways in which people practice religion? For better or for worse (I suspect for worse, but that’s a topic for a future email), the answer is yes, the world is already rife with instances of AI being used for religious purposes. Perhaps you’ve heard of clergy giving AI-assisted sermons. Companies are putting out glossy AI chatbots that claim to be assistants in interpreting Scripture. Then there are the frightening reports of AI-triggered psychosis with religious undertones.

And this is with the technologies that are already mature! Meanwhile, major tech companies are pushing for the development of “artificial general intelligence” (AGI)‒an AI that could, according to its proponents, do everything that humans could do, and potentially better. How deeply would this unsettle our notions of what it means to be human? After all, from a Christian standpoint, being human is to be made in God’s image.2 If we were to make an AI in our own image that shared many of our capabilities, what would set us apart from that AI? What is left of the imago Dei?

The Christian theological tradition, as ancient as it is, equips us with intellectual and spiritual tools to grapple with such difficult questions. Indeed, there might not be any satisfying way to address these issues without involving theological thinking. As you follow along with this newsletter, you’ll see me introduce a variety of Scripture passages and theologians who can help us reflect on these issues together. By subscribing to this free newsletter, you’ll be first to receive my updates!

Subscribe now