What Monks, Imams, and Skeptics Can Teach Us About Building AI

A few months ago, I watched a friend ask an AI chatbot whether she should leave her job. It gave her a tidy pros-and-cons list. It was helpful, in a way. But it couldn’t tell her anything about who she wanted to become — and that’s not a glitch to be fixed in the next model update. It’s a permanent limit. Decisions like that have always belonged to a different kind of thinking than the kind machines are good at.

This is where the world’s religious and ethical traditions earn their keep. Not because they have technical opinions about transformer architectures, but because they’ve spent millennia asking the questions AI can’t answer: What do we owe each other? What deserves our attention? What kind of life is worth living? If we’re going to build and live alongside increasingly powerful machines, we need more than engineering guidelines. We need a wider moral vocabulary — and these traditions hand us one, already tested across generations.

Here’s what four very different sources — Christianity, Islam, Buddhism, and secular humanism, plus one restless 20th-century monk — actually offer us.

Christianity: the person is not a product

Christian ethics keeps returning to one stubborn idea: every human being carries an irreducible dignity that no system gets to overrule. That has sharp implications for AI. A hiring algorithm that quietly filters out people by zip code, a chatbot that replaces a counselor instead of supporting one, a surveillance tool that treats a neighborhood like a dataset — these aren’t just policy problems. They’re failures to see the person in front of you as more than a record in a database.

The practical question this tradition pushes us toward isn’t “does this system work?” It’s “does this system still let us recognize the human being on the other end of it?”

Islam: You will answer for what you build

Islamic ethics frames human beings as khalifah — stewards entrusted with the world, not owners free to do whatever turns a profit. Nothing here is morally neutral, including code. A facial recognition tool sold to a regime, a lending algorithm that quietly redlines the poor, a weapons system that removes a human from the decision to kill — stewardship ethics insists someone is accountable for these outcomes, and that accountability doesn’t dissolve just because “the algorithm decided.”

This is a useful gut-check for anyone building or deploying AI: if you can’t explain who is responsible when it goes wrong, you haven’t finished designing it.

Buddhism: notice what you’re becoming attached to

Buddhist thought is less interested in rules and more interested in what’s happening inside us as we use these tools. Mindfulness practice asks: Is this technology reducing suffering, or just giving my craving a faster delivery mechanism? An infinite-scroll feed engineered to be unputdownable, a companion app designed to deepen loneliness rather than ease it, the low hum of anxiety that comes from checking a notification every four minutes — Buddhist ethics names these clearly as forms of attachment dressed up as convenience.

It’s a useful lens for both designers and users: not “is this technology good or bad” but “what is this doing to my attention, my craving, my peace of mind?”

Secular humanism: you don’t need a god to need ethics

It would be a mistake to assume only religious traditions have something to contribute here. Secular humanism arrives at strikingly similar conclusions through a completely different route — human reason, empathy, and our shared interest in flourishing, with no appeal to revelation or the sacred. Its core commitments are dignity, autonomy, fairness, and a refusal to let any person become a means to someone else’s end.

Humanist ethicists have been some of the most active voices in AI policy precisely because their framework doesn’t depend on theology to argue that an algorithm shouldn’t decide who deserves parole, or that consent matters when a system uses your data to shape your behavior. This tradition is a reminder that the conversation about AI ethics isn’t a religious-versus-secular divide — it’s a much larger, shared human conversation, and nonreligious people bring real moral weight to it.

Thomas Merton: the contemplative’s warning

I want to single out Thomas Merton, the Trappist monk and writer, because he saw this coming before most people had a name for it. Merton spent his life warning against what he called the “mass mind” — the way modern systems flatten people into roles, functions, and statistics, and the way a frantic, technologized life can drown out the quiet attention a person needs to actually know themselves or God.

Merton wasn’t anti-technology in some simple way. He was anti-distraction — wary of anything, including useful tools, that fills the silence we need in order to think clearly and love well. His questions land differently than the others above: not “is this just,” not “who’s accountable,” but “is this letting me become more fully myself, or just more efficient?” In an age of AI companions, AI productivity coaches, and AI everything, that’s not a small question.

So what do we actually do with this?

None of these traditions hands us a finished AI policy. What they hand us is a sharper set of questions — and, taken together, they give us a fuller diagnostic than any single one alone:

• Dignity (Christianity, humanism): Does this system still let me see the person, not just the data point?
• Accountability (Islam): If this goes wrong, who answers for it — and have we actually designed for that?
• Attention and craving (Buddhism): Is this reducing suffering, or manufacturing a new kind of hunger?
• Inner life (Merton): Is this helping me become more myself, or just more managed?
• Reason and fairness (humanism): Would this hold up if I had to justify it to someone with no stake in defending it?

No single tradition gets the last word, and that’s the point. The disagreements between them are as instructive as the overlaps. Taken together, they leave us with one closing question: will we let efficiency be the only question we ask about a powerful new tool?

Questions for discussion

1. Which of these five lenses — dignity, accountability, attention, inner life, or fairness — feels most absent from how AI gets built and sold today?
2. Merton worried about the “mass mind” decades before smartphones existed. Where do you see that warning showing up in your own relationship with technology right now?
3. Islamic stewardship ethics insist that someone is accountable even when “the algorithm decided.” Where in your work or community life have you seen accountability quietly disappear behind a piece of software?
4. Secular humanism reaches many of the same conclusions as religious traditions without appealing to the sacred. What does that convergence suggest about the source of these ethical intuitions?
5. If you could build one of these five questions into the design process of every new AI product, which would you choose — and what would actually change?