Back Navigation Menu
Search

Artificial intelligence and the future of religious thought: Compass 187

October 14, 2026

We're joined on Compass by DZ Kalman for a conversation about how artificial intelligence is reshaping our understanding of religion, ethics, and humanity itself.

Listen and subscribe: Apple Podcasts / Spotify / Amazon / YouTube 

We're featuring a conversation on how artificial intelligence is reshaping our understanding of religion, ethics, and humanity itself. D.Z. Kalman draws us into Jewish and interfaith perspectives on questions like: Can AI help us find new meaning in ancient texts? Does technology challenge our unique status as beings “created in the image of God”? How are faith communities adapting—and what moral boundaries should guide us as creators?

D.Z. Kalman is a leading thinker at the intersection of Judaism and technology. As a Research Fellow at the Shalom Hartman Institute, he directs a research group on Judaism and AI. He’s Senior Advisor at Sinai and Synapses and hosts the Belief in the Future podcast. D.Z. is passionate about helping religious communities navigate the ethical and theological questions posed by emerging technologies, drawing wisdom from both Jewish tradition and broader interfaith conversations.

Episode Notes:

Connect with DZ Kalman in a couple different places:

A few resources DZ mentioned during our conversation:

Some United Methodist-related resources on artificial intelligence:

In this episode:

00:00 Where are you using AI today?
05:21 Tower of Babel and Human Priorities
07:49 AI as a child metaphor
12:48 AI, theology, and ethical implications
14:27 AI and theological parallels
17:24 AI challenging human uniqueness
22:40 Centralized faiths and AI response
26:50 Ethical considerations in AI prompts
29:09 AI's potential long-term impact
34:02 Shifting focus to technology studies
36:25 Exploring spirituality and technology

Related Episodes:

Help us spread the word

  • Tell others: friends, coworkers, and anyone else might benefit from these conversations.
  • Share us on Facebook, Twitter, and other social media sites.
  • Review us on Apple Podcasts, Google Podcasts, Spotify, or wherever you download the episode. Great reviews help others find us.
  • Email our host Ryan Dunn about future topics and feedback.

More podcasts

Thank you for listening, downloading, and subscribing.

This episode posted on October 14, 2026

Episode Transcript:

Ryan Dunn [00:00:00]:
Do you think artificial intelligence has an impact on your faith? How about on your sense of what it means to be human? Yeah, we've got some deep topics that we're discussing with D.Z. Kalman on this episode of Compass. Welcome to Compass, finding spirituality in the everyday. In this episode, we sit down with D.Z. Kalman, a scholar and thinker at the intersection of faith and technology. Together, we're going to have a thought-provoking conversation about how artificial intelligence is reshaping our understanding of religion, ethics, and what it means to be human. D.Z. Kallman serves as a research fellow at the Shalom Hartman Institute, where he leads a research group on Judaism and artificial intelligence.

Ryan Dunn [00:00:49]:
He's senior advisor at Sinai and Synapses, assisting in developing Jewish pedagogical responses to technological development. And he hosts Belief in the Future, which is a podcast about religion, and technology, and it's my latest binge listen. It's excellent. DZ draws on both Jewish tradition and interfaith perspectives as we hit topics from using AI to translate dense religious texts and exploring the ethical dilemmas that arise, to grappling with theological questions about the image of God in the age of machine learning. Yeah, it's deep. Together, we examine how communities are adapting where spiritual language and policy fit into emerging technologies, and why the age-old questions of creation and morality feel more relevant and urgent than ever. So I hope you'll join us as we explore whether AI is just a tool or something that challenges the very boundaries of our spiritual realities. This conversation is good stuff.

Ryan Dunn [00:01:54]:
I really enjoyed it, even though I felt super underqualified to be a part of it. So thanks to Dee Zee for bringing this to my level. And let's get into it. DZ, thank you so much for joining us on the Compass Podcast. It's always pretty appropriate to start with a get-to-know-you kind of question, and I feel like maybe an appropriate question for the conversation that we're going to have is how has AI been a part of your day so far?

DZ Kalman [00:02:20]:
Like literally today?

Ryan Dunn [00:02:21]:
Yes.

DZ Kalman [00:02:23]:
Oh, you asked on the right day.

Ryan Dunn [00:02:24]:
Oh, great.

DZ Kalman [00:02:25]:
You really did. I have been fiddling around on linking up Claude, which is the AI that's my main driver, with OpenRouter, which is a more advanced tool that's used for allowing your AI to talk to a bunch of other kinds of models, other kinds of tools. And I've been using it to do something which seems kind of obvious, but I haven't seen anyone do before, which is to take major Jewish texts, major influential documents from the 20th century, Jewish philosophers, and convert those often very dense readings into 5 to 10 minute explainer videos that you'd be able to walk someone through in a very leisurely pace to make it a lot easier. I have been shocked by how good the results are. I've been shocked by how easy it is to take that content and make it accessible. And it is so easy that it is making me ask like all these questions about like, how is it appropriate to use this? Because I can see like, I know I'm early on this, but I know that in 3 months everyone's going to be in the same place as I am right now. And I am worried about that flood of translated religious, often like very difficult religious into material that is just everywhere and seems so easy that it feels like, what's even the point of making this? Because it's everywhere already.

Ryan Dunn [00:03:41]:
Right.

Speaker C [00:03:42]:
And anybody can be doing it on your own. And isn't that kind of the paradox of this time in which we sit in? You've done this thing that you're incredibly excited about and interested in, and yet shocked in a way that is almost a little bit scary, right? There is a fear of what the future might look like in what you're doing, correct?

DZ Kalman [00:04:04]:
And it also feels like there's this triangulation which I don't know, like, I'm old enough to remember the days of Napster and the kind of internal debates that many of us had around, like, what is the ethics of pirating this Metallica song?

Speaker C [00:04:15]:
Yeah.

DZ Kalman [00:04:15]:
Where you'd have this conversation in your head where you'd be like, well, but I was never actually going to buy that album, right? So, like, of course it's fine. Um, and I'm having the same conversations, like, I was never actually going to hire that graphic designer, so maybe it's okay. But of course, like, the presence of the tools influences the decisions that we make and where we choose to put our resources. So how honest can we really be in those conversations? Yeah,

Speaker C [00:04:39]:
well, you talked about using the AI tool to kind of distill down some meaning or logic behind a deep theological text. And I wanna go ahead and jump in and get a little biblical, if we may. I'm gonna quote you from 6 months ago, and I do so noting that you've also said in other places that a lot of the statements that we're making right now need to just be like 1-month statements. We're not making statements for all time. But you did call the Tower of Babel one of the most important biblical passages for the moment, at least the moment 6 months ago. Can you talk about why this story from our heritage is so important?

DZ Kalman [00:05:21]:
Yeah, so the Tower of Babel story, right? Like, just to give a quick recap of it, you have a bunch of people after the flood, all of humanity, kind of come to this valley and decide to build a tower. It's unclear exactly why they're building the tower. It's also unclear why God is unhappy with the building of this tower, but they start building the tower. God disrupts this building process by making people unable to understand each other, basically inventing multiple languages, and then they kind of disperse and spread to the rest of the world. In the modern context, it's not hard to see why this text is picked up as being about technology and being about, you know, societal project and about who gets to speak for humanity, right? There's this amazing Midrash, right? Like a kind of early Jewish form of commentary by the rabbis where they talk about the problem being, if you were on the Tower of Babel and a person falls off the side, they'd be like, okay, we'll get another person. But if a tool falls off the side, they'll be like, oh my God, when will we find another one? And they would mourn the tools more than they would mourn people. Mm-hmm. And that being the kind of instigation for God to come down and say, okay, this is enough.

DZ Kalman [00:06:25]:
You're pursuing a project that is clearly putting something other than human beings at the forefront. And so that is potentially dangerous. So that seems like a pretty easy to go— that's a pretty easy metaphor to use in thinking about AI. You can go all kinds of places with it, but it feels like it jumps off the page when you read the text today.

Speaker C [00:06:43]:
Yeah, well, another line of thinking along the Tower of Babel goes into this idea that in some way they're building this thing to rival God, so to speak. And it's kind of interesting to look at the development of AI as a parallel for how religion has, in a sense, developed over the years in terms of how we think of of human beings as being, in a sense, children of God. And we called AI a child of the human creation. In some ways, like, we have become the creators of something that, in a sense, could overtake our understanding or ability to control the created, which, I don't know, maybe is a parallel for the way that humanity moves through our own creative story, correct? Yeah. That we have this sense of being able to replace the creator or a fear that we might.

DZ Kalman [00:07:49]:
Yeah. Look, I find the child metaphor very useful for a few reasons. One, because it implies a kind of responsibility, a kind of open-ended responsibility, right? You don't know what kind of child you will get when you get a child, and yet you are responsible for them however it is that they develop. And that seems appropriate to AI. It also seems, in my conversations with folks at Anthropic and in AI safety companies, the language of this as being a product that is grown as opposed to being a product that is just built shows up over and over again. I think it's a language that seems to occur to people spontaneously all over the place. And of course, the idea that a child grows up and can become more powerful than you and can surpass you in all kinds of ways. But the thing that I like about the metaphor the most is that by talking about AI as a child, you are talking about a process that exists in cycles, right? You were a child, and now you have a child, and then the child has its own child.

DZ Kalman [00:08:47]:
Yeah. And the sense of AI as a kind of recapitulation of a process that humanity itself has undergone, and there you can go back to biblical text and thinking about humanity as God's child or as God's progeny, I think is very helpful in thinking about the moment we're in and the position we're in. And it gets you out of that dilemma that I already hear you starting to ask, which is like, where does God language fit in here? Because I don't want to be calling AI a god. I think that's dangerous. I think that's pretty non-controversial. I also don't know that AI is exactly a human being. AI is its own thing. We have never seen anything like AI.

DZ Kalman [00:09:23]:
And I think we're kind of slowly coming around to the idea like, oh wait, this is its own thing that does not have a category that we have used previously. But it is helpful in thinking about ourselves not as being gods, but being in the place that God perhaps sat. while creating humanity and all of the various dilemmas that go into what is it to mean, what does it mean to create something that both has its own free will and also needs to be governed according to laws and all the many ways which, if you read the Bible, human beings are constantly going off the rails regardless of how many laws they're given.

Speaker C [00:09:56]:
Yeah. Do you think that religious language has a place then in helping to define like what the constituting documents or manifestos of AI might be?

DZ Kalman [00:10:08]:
Yeah. So let's talk about religious language in 2 senses. There's small r and big R religious language. So the small r religious language is everywhere. If you talk to folks in Silicon Valley, they are evoking religious language even without being affiliated or wanting to be affiliated publicly with any particular faith, because AI gets you to ask all these questions about the nature of humanity and our place in the universe, and what does it mean to create something that is this momentous? So it shows up, and no one has a copyright on that language. Everyone gets to use it. Now, brand name religions, like capital R religions, they use that language because people want to ask those questions. And they've developed over centuries, over millennia, very sophisticated responses to thinking about those things.

DZ Kalman [00:10:53]:
And so there's both the small r religious sentiment that you see, and also religions as organizations, as communities, trying to resolve those questions together. For the second piece, something that I think is really important is to recognize that religious communities don't just exist as stores of knowledge. They do exist as stores of knowledge, But they also exist as communities, as people who come together to think through what does it mean to live a good life and how to be in the world, and what does it mean to raise children and all those other things. So the presence of those communities asking those questions right now, I think, is particularly important because we actually, as a society, don't have that many mechanisms for making moral decisions.

Speaker C [00:11:31]:
In this interplay between, I guess, AI and religion, you've noted how AI has helped you to discern a little bit about your religious tradition and looking through previous teachings and being able to distill those out. I don't know, maybe it was able to distill something out for the less initiated viewer or reader, and it was things that you already knew, DZ. But do you think that AI has, in a sense, impacted your faith?

DZ Kalman [00:12:04]:
Yes, it definitely has. I feel like we're in a— there are certain moments in history when I think a lot of people have a sense of like, there is some turning point happening here. And I think people have that sense of this being one of those turning points, and it leads to all kinds of behaviors. But I think like, look, just kind of speaking from within the perspective of Judaism. So I think Judaism has struggled with theology for a long time. It's struggled especially since the Holocaust. Out of this question of, like, how do you find a theology that actually matches the experience of people in your community? And I think there is a whole world of post-Holocaust theology. I think it has provided perhaps technical answers.

DZ Kalman [00:12:48]:
I don't know that it's provided answers that people always feel on an emotional level. And so Judaism in some ways is a religion, and this is like my editorializing, is a religion that is seeking new forms of theology. Yeah. And I think there is something in AI and the way that we are encountering AI, thinking about what does it mean to create covenants with this technology? What does it mean for it to be good? What does it mean for us to be good stewards of the technology that are important in allowing us to revisit our own perhaps fractured relationship with our own creator and to see that relationship and the complexities of it from the creator's perspective and how the things that we perhaps want from our creator are not always— so easy to come by because despite the fact that there are these essential similarities between God and humans and also humans and AI, there are also ways in which we can never fully understand each other because on a substrate level, we are just fundamentally different.

Speaker C [00:13:48]:
It sounds almost like you are entertaining the idea of using AI as a way of role-playing in a place of God. Is there a way that practically that you've engaged in something like

DZ Kalman [00:14:03]:
I think roleplay is a good word for it. Look, you know, within the Hebrew Bible, you have this idea that human beings are created in the image of God, right? Imitatio Dei in the Christian tradition. There's a Midrash that makes this incredible point that, you know, we often think about being created in the image of God as being about like, well, mostly we're not God, but like there's this little aspect of ourselves, like some spark that is godlike.

Speaker C [00:14:26]:
Yeah, sweet.

DZ Kalman [00:14:27]:
There's a Midrash that that kind of flips it on its head, where it says when human beings were created, when Adam was created, Adam was so similar to God that the angels actually got confused. The angels started worshiping Adam until God had to be like, hello, I'm over here. That's Adam, that's not me. And that confusion resonates with me quite a bit in the AI era, where we have these constant questions about Well, but AI is doing all the things that we thought were unique to us, and how do we make those differentiations? And the kind of like the uncomfortability of that resemblance seems really important for the present moment. So is it role play? Maybe it's like play play. I don't really know. But like in one way or the other, like I think we are being given the opportunity to go back to theology as a kind of precedent for the present moment, which I don't know that we ever thought was going to happen because why would you think about theology as a precedent? What would it be a precedent for? But it turns out it's a precedent for this, I think. I think it really does help in imagining what does it mean to be in that role and to be filling those shoes.

Speaker C [00:15:36]:
Can you tell us a little bit more about being created in the image of God? You've noted that whenever sort of religion intersects the AI world, it's only a few sentences before somebody says something about like being created in the image of God. And I guess that's because AI in a sense challenges our assumptions about what it means to be created in the image of God because it can recreate so well our human characteristics. Are there other ways that AI kind of challenges this idea of the imago Dei or the imitatio Dei or the image of God? Yeah, so there's an incredible

DZ Kalman [00:16:15]:
book by Tomer Persico about this, which I have learned a lot from, which is not about AI, but feels so relevant for the AI moment. Tomer talks about how the notion of the image of God, of human beings being created in the image of God, is something that is kind of co-created between Jews and Christians, where it exists originally within the Hebrew Bible as a way of saying that human beings are kind of the ultimate unit of morality, meaning morality doesn't is not something for a household or for a town where any individual human beings can be collateral damage and all you really care about is the welfare of the town. But instead, every human being is an end in and of itself, to borrow some anachronistic Kantian terminology. But every human being matters in and of themselves. So that starts to show up. It evolves really quite prominently within the Christian tradition as well. One of the things that gets picked up in the medieval period is this idea that we want to identify what specifically within human beings is the image of God part? Like, is it the free will? Yeah. Is it agency? Is it some kind of rational thought? Is it the ability to speak? That all makes sense.

DZ Kalman [00:17:24]:
It especially makes sense at a period when there is nothing really quite like human beings that you might suggest, well, what about this thing? Doesn't this also have the image of God in it? One of the things that's kind of backfired in the AI era is that the more that we connect being created in the image of God to some unique aspect of humanity, the more it ends up being threatened when AI takes away those things that are unique. So we thought that we could be special because we had this long list of unique characteristics. And every year, every month, every week, AI makes that list shorter and shorter and shorter. And so we start questioning, do we really deserve that uniqueness? So what's happening now is we're kind of shifting from, well, we matter because we're unique, to we matter because we matter. We're not going to rationalize it. We matter because axiomatically, we've decided that human beings are important. And I think that shift is easy for some people and it is not so easy for other people. I think it is simpler to imagine that human beings matter because they do have special characteristics.

DZ Kalman [00:18:26]:
And it is going to be hard in practice when AI is so prominent in the world to really remember why it is that human beings are valued over beings that are far more capable.

Speaker C [00:18:38]:
It's mind-blowing in a way to think about. I guess when we tried to distill, like, what is it that we are that makes us uniquely made in the image of God? And you've talked about how in our Christian tradition, we tend to lean into the idea of being incarnate and physical. And in the Jewish, because in the Jewish tradition, like, that's not quite as present, that there's something else there that's within us. It might be the ability to reason, the ability to empathize, the ability to create and have agency. As you've noted, do you think that in a way the image of God might be a sense of that innate moral character that we carry? And is AI able to replicate that? Like, have you seen ways that AI moralizes?

DZ Kalman [00:19:36]:
It's interesting. I tend to kind of stick with the Midrash and think about the image of God as being a kind of uncomfortable resemblance. And it's not any one thing. It is, in fact, many things, perhaps too many things.

Speaker C [00:19:51]:
Yeah.

DZ Kalman [00:19:52]:
And all of the dilemmas that come about come about because of that uncomfortable resemblance. But the question of where does morality fit into this, of course, matters quite a bit because we imagine our morality within many religious traditions as descending from a divine standard. we give AI morality based on our own standards. One of the things that is so tricky is that even in those conversations that we so often have about how appropriate is it to anthropomorphize AI, we don't really have language other than our own moral language to supply AI with ideas about what is right and wrong. It's not like we have some special non-anthropomorphic morality that we can hand over to AI. We end up needing to explain to AI what is goodness, what is love, what is compassion. It ends up showing up regardless. And it's part of the reason why I am skeptical that we will be able to really say definitively that AI has no moral status.

DZ Kalman [00:20:51]:
Those resemblances are just going to show up so frequently because they're part of the development process and they're part of the use process.

Speaker C [00:20:59]:
Okay. Is that because it's, I guess, in terms of the model that it's learning from our own human history and human contributions, and so it just has that base understanding of?

DZ Kalman [00:21:12]:
Yeah, it's both that, but also it's, you know, look, there are non-anthropomorphic AI models out there, right? So the AI that beat Garry Kasparov at chess is not an anthropomorphic AI. The AI that folds proteins is not anthropomorphic. The AI that wins Go championships is not anthropomorphic. But the AIs that we use, the ones that are kind of these general purpose models, they very much are given personalities, they're developed using personalities. That is an important piece of how they exist in the world. And so we have decided that it actually matters quite a bit to have this resemblance be a feature of the development process, because it ends up affecting how the AI models exist in the world, and it makes it easier for us as human beings to control them. We have a much easier time controlling something that is in the shape of a human than something that that has a completely alien shape.

Speaker C [00:22:12]:
Yeah. Okay. So in a way, it's just for our own comfort that we tend to anthropomorphize. Interesting. Well, in your experience in, I guess, going to different conferences, meeting in different educational spaces, and even through your own podcast, you've been able to speak with a number of different faith leaders from different traditions. What are some ways that different faith groups are responding to the rise of AI?

DZ Kalman [00:22:40]:
So this is like a big dumb generalization, but I do think there is like a— there's an important division between centralized and decentralized faiths right now. Yeah. So centralized faiths are able to create responses to AI more quickly. Sometimes they can create policies around AI more quickly, both because they have mechanisms for making decisions and for putting out like, you know, big statements that represent the denomination. But also, and this is like, I think, not so well appreciated because understanding AI well enough to talk about it is still specialized knowledge, right? A lot of people understand it to some degree, but understand it to have the confidence to really speak about it is specialized knowledge. And if you're a centralized faith, you can say, great, we're going to find the people and we're going to put them in a room together and we're going to have them talk and we're going to have them release a statement. That is a luxury of, say, the Catholic Church, right? Yeah. So the encyclical, which is an incredible document of exactly that type.

DZ Kalman [00:23:37]:
And also the LDS Church, the Mormon Church, has similarly kind of put out both grand theological statements and also created policy statements. Decentralized faiths struggle much more because in theory, what happens is each church, each synagogue has the ability to make its own choices about how AI shows up or how you talk about AI. In practice, if you are a local rabbi, you do not have time in your day to be pontificating about AI, right? You have a day job, right? Yeah. And maybe you have some predilection towards it. You know, maybe you're the the kind of rabbi who grew up reading science fiction, and so it interests you, and so you end up talking about it naturally. And I think a lot of the times in decentralized faith, that's exactly how ideas about AI percolate upwards. But a lot of the times, it's just kind of out of your reach. And so I see decentralized faiths, both large faiths like Islam and also small faiths like Judaism, really struggle with how do you formulate responses.

DZ Kalman [00:24:38]:
So this is a problem that I'm trying to tackle right now. How do you create kind of model policies that can be adopted more easily to kind of like do some of the heavy lifting for individual congregations or communities or organizations. But that's a major division that I see right now in the developing AI religious responses.

Speaker C [00:24:57]:
So you are working in sort of policy development?

DZ Kalman [00:25:01]:
Yeah, stay tuned in the next few months. But I'm working with a team at the Shalom Hartman Institute. I have a team that works on Judaism and artificial intelligence that's trying to figure out like, how do you create model policies for the use of AI, understanding, I think this is also a crucial piece of it, that no one is able to make a policy that will last even a year. The technology is moving too fast. So part of what it means to create guidelines is to know that they're only going to be relevant for a few months, be okay with that, and be willing to revisit them as new needs arise, as new capabilities emerge.

Speaker C [00:25:40]:
All right. So just out of curiosity and understanding that you haven't rolled anything out yet, do you have a couple sort of key factors that might be rolled into a policy? Or even if you're speaking to a local religious leader right now, like what might be one or two things that they want to implement in their own AI usage policy?

DZ Kalman [00:26:06]:
Yeah. So I don't want to, you know, like put the cart before the horse because we haven't finalized anything yet. Definitely, yeah. But one thing I will say, and this like comes also deeply from the Jewish tradition, is there's a rabbinic text, a core rabbinic text called the Mishnah, which is a set of laws. One thing that people who encounter the Mishnah for the first time find surprising is it is not programmatic. It doesn't usually start, like each individual tract, it does not start with like, here's all the big things you need to know and here's the sub-things. It'll often start feeling like you're already in the conversation and it's just kind of continuing on with more examples, more examples. But what ends up happening is that by getting a number of examples of like, what do I do in this situation? What do I do in that situation? You kind of pick up a feel of what is okay and what is not okay.

DZ Kalman [00:26:50]:
I'm so curious about how that intuition develops around AI and specifically around which prompts are ethical and which ones are not. Right. So if I tell an AI, speaking as someone, as a pastor of a congregation, know, write me a sermon, pretend you're me, press enter, right? I think most of us would have the intuition like, that's not okay. Yeah. Right? Whereas if I said, you know, here is my completed sermon, please list all the grammatical and the spelling errors, you would say that's fine. But articulating and giving people a kind of clear sense of like what kind of things are reasonable to ask AI, what kinds of things are unreasonable? What kinds of things are reasonable to ask, but only in moderation? Because if I overuse that, then I will lose skills which I've spent a long time developing. Those are all important, right? Like, AI is a great translator. AI can translate very difficult medieval Hebrew that's full of all kinds of scriptural references, often flawlessly or like 99% accurate.

DZ Kalman [00:27:57]:
If you are the kind of person who spent a lot of time developing the skill to be able to do that yourself, and now AI can suddenly do it with much more ease, Like, you have to make a decision. Is this a skill that you want to retain for yourself, or do you want your students to pick up, or is it okay to kind of let it go? So those are the kinds of things. But I think focusing on like, what are ethical prompts and what are unethical prompts is a useful way to start thinking about the ethics of this.

Speaker C [00:28:23]:
As an ethical religious person yourself, has there been an instance when you've been engaged in something and exploring AI usage and you've said, oh, I need to back off of this?

DZ Kalman [00:28:37]:
Yeah, I mean, like, in general, the place that makes me feel really uncomfortable is asking AI to write in my voice. Mm-hmm. Partially that's because I like writing, and it feels like maybe I can do this, but I don't want to know. But in general, like, I am still trying to figure out what the rules are, like what the guidelines are, especially in situations where it seems so easy to just like, you know, just do it. Like, forget about building a team, forget about like, you know, human expertise, just like do it yourself.

Speaker C [00:29:09]:
Yeah.

DZ Kalman [00:29:09]:
So, you know, I've struggled with that. At the same time, I'll just say like an important caveat for all of this is AI can be an addictive technology. Addictions do not develop at the same pace. Some addictions develop very quickly, some addictions develop over years. And it has only been a few years since we've all started using AI. And I don't know that we can fully see yet all of the long-term impacts of the way this technology will affect us. So even if a certain use feels safe today, that use over the span of a decade may be dangerous. And I think like we are learning through, you know, crazy trial and experiment, trial and error right now.

DZ Kalman [00:29:51]:
Exactly.

Speaker C [00:29:52]:
In your work across different religious institutions, have there been some ways that you've noticed AI harming those institutions?

DZ Kalman [00:30:01]:
I get particularly nervous when AI is used as a replica of living wisdom. I don't love seeing pastors replicate themselves and tell congregants to talk to the replica instead. I also worry about turning yourself into a chatbot or things like that. Look, those kind of uses do have a place. Religious questions, like questions about sexuality, I think are the kinds of material that people often do start with the internet for because they feel, you know, they're private. Yes. And, you know, you want to start without interacting with another human being. So I expect that, you know, a lot of religious inquiries are going to start with a conversation with AI.

DZ Kalman [00:30:56]:
But I think the responsible thing to do is is for AI systems to kind of push people gently towards interacting with other human beings, understanding religion as something that often happens between other human beings in physical spaces. And I see a lot of the kind of the purpose-made chatbots from many Christian denominations, from Muslims, from Jews, often have that as part of their programming, this sense of like, okay, I'm glad you're here, but maybe you should talk to a rabbi, maybe you should talk to an imam.

Speaker C [00:31:26]:
Yeah. Okay. Yeah. And even to maybe pull back the veil a little bit more, one of the ideas that we've played around with within our communications agency is how can we create a chatbot where people can ask questions of voices of our religious traditions past? And it goes into that area that you're kind of worrying about where, well, are we just recreating this voice, which may be unfaithful to the tradition, or can we just put it in a way where it's like, it's almost like a library reference and the AI becomes the librarian in being able to present, well, this is where this person said, historically speaking, some kind of answer or something relevant to the query that you're bringing to us. Right.

DZ Kalman [00:32:15]:
Yeah.

Speaker C [00:32:16]:
That's good. Are there ways that, maybe I just lifted one up, but are there other ways that you've seen AI helping religious institutions?

DZ Kalman [00:32:24]:
I mean, look, on the backend, you know, in terms of finance, in terms of like writing grants, things like that, sure, it's useful in any number of ways. Like it can save time. It can save time around the edges. And as a research tool, you know, I have found materials more quickly, more efficiently with AI than I would have found otherwise. You know, within Judaism, there's a a tool that really just launched a few months ago called Yochai, that when you ask it to give you answers on questions of Judaism, asking about particular texts, it will try to engage you in a kind of Socratic dialogue rather than just giving you an answer, like trying to replicate the experience of being in a study hall of actually needing to put a lot of work on you, which I like as a model. So I think it's evolving. I think it's developing. I have no doubt that religions and communities will use AI in all kinds of interesting ways.

DZ Kalman [00:33:28]:
I don't know that I've seen so many ways that are unique to religious institutions just yet. They feel like the kind of like the religious version of what other people are doing in other domains, but, you know, they're still valuable.

Speaker C [00:33:41]:
Well, DZ, I've had a great time kind of going down the rabbit hole of your past explorations, and I am a little bit curious on how you even got into this whole intersection of AI and religiosity. Do you have a moment to kind of explain how you've gotten to where you are right now?

DZ Kalman [00:34:02]:
Yeah, so I grew up in an Orthodox Jewish community, went to Jewish day schools, went to yeshiva afterwards, became interested in Jewish law and the relationship between law and religion, and started studying other religious traditions, started studying Islamic law, started studying Arabic, and spent a long time in grad school trying to think about the relationship between Jewish and Islamic law. At some point, the questions of technology just kind of became impossible to ignore. And with some advice from my wife, really shifted the course of my studies towards technology. Interesting. And starting to think about how do these things connect? There really isn't a field of Judaism and technology studies. It does pop up in places here and there, but Judaism in particular has a complicated relationship technology because technology is often connected to power and authority. And for most of Jewish history, Jews did not have power or authority. So there aren't that many Jewish writings about technology, at least up until like the last 500 years and even more the last couple of 200 years.

DZ Kalman [00:35:09]:
So, you know, I kind of came to it because it seemed kind of unavoidable. And then, you know, around 2022 or so, the world started to pay attention to this stuff too. So it's been like a strange journey in going from something that felt very niche to articles every day, every other day about relationship between AI and faith.

Speaker C [00:35:34]:
Yeah. Okay. So it's a situation where once upon a time you might have to spend 5 minutes explaining what you do and now you can kind of sum it up in 14 seconds or so. People just kind of get it. Well, I know that probably a great place for people to follow along with your writings is your Jello Menorah website. I would love to know the explanation behind that name. Is that public knowledge?

DZ Kalman [00:35:58]:
Um, it is meant to be an exercise for the imagination.

Speaker C [00:36:02]:
Yeah. Okay.

DZ Kalman [00:36:03]:
Um, but I, you know, I look, I, you know, as separate from my work on AI, um, I like to engage in art and religious art and like the idea just, just tickled me and it sticks, it sticks in the mind and that's really what it was intended for.

Speaker C [00:36:15]:
It does. Perfect. Okay. DZ, thank you so much for sharing your time and your experiences with us. It's much appreciated.

DZ Kalman [00:36:23]:
Great speaking with you too.

Ryan Dunn [00:36:25]:
Thanks for joining us on this technical, challenging, hopefully hopeful trip on Compass, Finding Spirituality in the Everyday. If you'd like to dig deeper, head over to our website, umc.org/compass. There you'll find episode notes, more episodes to explore. We have more to say on AI and on the intersection of technology and spirituality. Tech and Faith. But if you want to take the real deep dive, check out D.Z. Coleman's podcast, Belief in the Future. You can also find so much of his work at jellomenorah.com.

Ryan Dunn [00:37:01]:
And a big thank you to the team at United Methodist Communications for making this podcast possible. If you haven't already, make sure to subscribe to Compass on your favorite podcast platform. And if you're enjoying the show, Please take a moment to rate and review. Helps others find us and supports the work that we're doing. Thanks again for listening. We'll see you again in 2 weeks' time here on Compass. Peace.


Related Tags

Back to top

United Methodist Communications is an agency of The United Methodist Church

©2026 United Methodist Communications. All Rights Reserved