Angela Heck (00:00:04):
Hi, and welcome to the Ask the Pastors podcast, where our pastors here at West Hills Church answer your questions for the benefit of the church. My name is Angela Heck, and I’m excited to be your host this season.
Will DuVal (00:00:14):
Heck yeah, you are. I can make that joke at least a couple more times.
Thad Yessa (00:00:19):
Every episode.
Will DuVal (00:00:20):
Yeah.
Angela Heck (00:00:21):
This week we are talking about AI and personhood.
Will DuVal (00:00:30):
I figured since we’re talking about what it means to be human, it was appropriate for me to eat my lunch during the recording of this podcast since machines can’t do that. I mean, they can probably. AI can’t though, right?
Angela Heck (00:00:46):
Will is not AI. Confirm.
Thad Yessa (00:00:49):
He’s confirming he’s actually recording this podcast. Yeah,
Will DuVal (00:00:51):
We’re keeping it human. Sorry, go ahead.
Angela Heck (00:00:54):
So our question for this week is what does it mean to be human in the age of AI? Could AI ever develop personhood?
Will DuVal (00:01:04):
What do you think, Thad? Would it be helpful to start by defining some terms?
Thad Yessa (00:01:11):
Probably.
Will DuVal (00:01:15):
Personhood, the state of being a person. A person, according to the dictionary, as a human being is distinguished from an animal or a thing, or a subsequent definition in the branch of philosophy is a self-conscious or rational being. I think words that often come up in this discussion are consciousness, the state of being conscious, sentient, having the power of perception by the senses, conscience. Anyway, I wanted to, before we even dove into the personhood question, I feel like there were a couple questions from last week that we didn’t get to finish. Do you want to pick up there where we left off? Mostly
Thad Yessa (00:02:11):
Talking about what we as pastors use AI for, don’t use it for, how should our people in the church?
Will DuVal (00:02:17):
How the church should, both with their work and in church capacities and… Well,
Thad Yessa (00:02:22):
I think it’s wrapped up both in personhood, talking about our work, our vocation, as well as maybe it’s next week we’re talking about specifically in our work and vocation, a theology of vocation. Maybe I’m wrong, maybe I’m off script.
Will DuVal (00:02:41):
Maybe. Okay. So you want to hold off on some of that?
Thad Yessa (00:02:44):
No, we can. Well,
Will DuVal (00:02:49):
I don’t know. I was thinking more about…
Thad Yessa (00:02:53):
I’ve not thought about what we recorded last week until this moment when you just asked me.
Will DuVal (00:02:57):
Okay. So let me ask you this. This is not a hypothetical. I had one of our small group leaders at the church co-lead together as a couple. And after I announced last week that we’re going to be restarting the podcast and specifically talking about AI the first couple weeks, she mentioned that when, we’ll remain unnamed, when her husband is out of town midweek traveling for work, that she jokes with her group. It’s like their ongoing joke now that she’ll be co-leading this week with her new co-leader, Claude, and that she’s using Claude and AI to give summaries of the passage for the week that they’re studying, help write discussion questions.
(00:04:06):
I don’t want to over speak and say things that aren’t true in terms of exactly how she’s using it, but suffice it to say, sounds like using it, maybe leaning on it even pretty heavily for facilitating small group. Is that something that as a church we should be concerned about? Is that something that we should step in and say, “Hey, we need an AI usage policy for whenever you’re doing anything for, with, through the church, functioning in a church capacity, would it be an overstep if we as a church leadership put out a policy that not only extended to how we’re going to regulate expectations for AI usage when it comes to church functions, but I guess you probably wouldn’t regulate, but for us to put out a policy that would give direction and guidance to our members for personal usage of AI or even professional, some of that…
(00:05:37):
Again, right now, because right now we’re just speaking very informally, sort of ex cathedra as the Roman Catholics would say when the pope’s just kind of shooting from the hip.
Thad Yessa (00:05:50):
Just teasing a little bit.
Will DuVal (00:05:53):
It’s
Thad Yessa (00:05:53):
Not policy.speaking
Will DuVal (00:05:54):
In his official papal sort of authority. This is not me standing up at the pulpit. This is not us putting something in writing. But anyway, should we? Hearing that we have at least one small group leader, she’ll for now remain nameless, using AI in that kind of way. No, I’m getting
Thad Yessa (00:06:15):
Anywhere.
Will DuVal (00:06:16):
Oh, you know.
Thad Yessa (00:06:20):
I know of one other individual who was using it in a D group context to help write Bible study curriculum in one sense of here’s what we’re walking through, here’s the resources that it’s pulling from, here’s a copy, would you like to look over it? What do you think of it? Is this helpful? Stuff like that. Curriculum writing, a little bit perhaps different. I would say of our shooting from my small group leader, you should not need Claude to facilitate conversation in your small group. If you serve as a leader of our life group, you should have the biblical maturity and understanding to be able to… We as a group are going to read through the text and we can talk and apply. I do not think Claude, ChatGPT, whatever, should be doing your biblical interpretation for you in an authoritative manner.
Will DuVal (00:07:23):
So if I was going to play devil’s advocate and say, if I was this leader and said, “Well, I don’t need it. I could facilitate small group without it, but is it a helpful, useful tool? Is discussion going to be better if I use it? I’ve done it without it, I’ve done it with it, and it seems like sometimes it’ll come up with a question I wouldn’t have thought of or whatever. And so maybe it’s just a tool that augments, just like you don’t need a dishwasher to clean your dishes.
Thad Yessa (00:08:05):
Facts. Currently don’t have one.
Will DuVal (00:08:07):
However, is it helpful? I mean, is it a helpful tool that helps you get it done better quicker?
Thad Yessa (00:08:14):
Yeah, I think my response to that would be where we were last week of what’s the purpose of small group and does AI care about your soul? The answer is no. And so again, my confidence in humanity to have even a fellow Christian member of our church, I would say it is probably more unhelpful for you to rely on that as a consistent crutch versus developing… We just had an apple explode in studio. Well,
Will DuVal (00:08:52):
We’re going to play the game of guess what he’s eating, but now I blew it. I literally blew up the apple.
Thad Yessa (00:08:59):
So I would say, again, if we’re going to view it as a tool, I would love to see that individual become less reliant on it and develop the skill of asking good questions. I think that is a skill that has been lost over the last couple years as we’ve become less literate of people and AI is going to cause us to become even more illiterate. And so to develop that skill as well as to not… If I’m going to say, “Yeah, you can use it, but here’s my caution,” I would say use it, but I would not just take every question as a question worth asking because I could ask right now, “I got a manuscript from Jake’s teaching at youth group tonight. Here’s what he’s teaching on ChatGPT. Give me 10 questions that our small groups can ask.” A good leader should not just say, “These are the questions and we’re going to do that.” It’s going to sort through what’s good, what’s helpful and not.
(00:10:00):
And so to, again, even as a tool, not just take it as a tool for this is what it’s worth, but it still needs to be evaluated and sorted through and really tested before just saying, “Well, here’s the questions for the group.”
Will DuVal (00:10:15):
Yeah.
Angela Heck (00:10:16):
I have a question. Are we saying that asking ChatGPT or whatever to form questions about biblical text is biblical interpretation in a way?
Thad Yessa (00:10:28):
Yes. I think you’re asking it to do biblical interpretation.
Angela Heck (00:10:32):
Yeah.
Thad Yessa (00:10:33):
And now it’s a different thing if you’re wanting to
Will DuVal (00:10:40):
Know – Simply asking questions in and of itself qualifies as interpretation.
Thad Yessa (00:10:47):
If you’re asking, what is the meaning of this text?
Will DuVal (00:10:50):
Oh, if it’s that specific question. If you’re asking it that question, if you’re asking… But you’re saying if you’re using AI to develop questions.
Angela Heck (00:10:59):
Yeah, like small group setting, what questions could I ask about this text?
Will DuVal (00:11:05):
Is inquiry a form of interpretation in and of itself? That’s an interesting question. I don’t know. Well –
Thad Yessa (00:11:15):
There’s an interpretative measure that’s taken into the question it’s
Will DuVal (00:11:19):
Going to ask.
Angela Heck (00:11:22):
Yeah. Because you could ask not right questions if your theology is not looking at the text correctly.
Will DuVal (00:11:29):
Is there harm in asking
Thad Yessa (00:11:30):
Questions? I think, again, I’ll just say a general concern, even if you put… I know that the Gospel Coalition at the women’s conference, it’s got to be within the Westminster catechism and London confession, stuff like that.
Will DuVal (00:11:47):
Yeah. We mentioned that last week, right?
Thad Yessa (00:11:49):
If you’re putting in certain filters, you might get really good sound questions, but again, I’m assuming our people are not saying, “Do this.” I could be wrong. But
Will DuVal (00:12:05):
You should.
Thad Yessa (00:12:06):
But you should.
Will DuVal (00:12:07):
Yeah. I mean, if you’re going to use
Thad Yessa (00:12:08):
It – If you’re going to use it for any sort of biblical interpretation application, you’re adding –
Will DuVal (00:12:14):
Prompt engineering to your prompts. So that – Give your answer in the theology of the Gospel Coalition. Give your answer, make your response consistent with the Westminster Confession of Faith or London Baptist 16, whatever.
Thad Yessa (00:12:34):
I have one other just piece to it. Again, I’m not saying I’m a source of all knowledge or that Will’s a source of all knowledge or that our elders are a source of all knowledge. I would probably prefer someone to text me and ask me, “What are your thoughts on this text versus going to AI for it as their only source or not doing, again, what some might consider the hard work of what does my study Bible say about this?” Again, a more trusted source than OpenAI is if you’re not going to put in all the different in the theology of in alignment with categories.
Will DuVal (00:13:12):
Yeah. Yeah. I mean, I might as well be… And here’s what I would say. And this is what I wanted to say about it, and both the small group leader kind of responding to that as well as just at a macro level. And we talked about this last week, I know with cooking and needing mom’s recipe, do you pick up the phone and call her or Gemini? But the reason that I would prefer maybe someone to text me and ask me, “Hey, what do you think about tattoos or whatever it is, is not because I’ll be honest, I’m probably going to use the internet to help me remember exactly where in Leviticus it talks about tattoos. I’m going to do that myself anyway in my response, but just because I don’t have the whole thing memorized and ChatGPT does. And so it’s a powerful tool.
(00:14:19):
But again, it’s the connection. So now you’ve given yourself a chance to connect with me, your pastor, you’ve given me a chance to connect with you congregant in my church that I’m pastoring. And it’s that human connection piece that is important. Just like mentioned last week, my dad texted me about who won the Heisman Trophy in 1993 or whatever.
Thad Yessa (00:14:47):
Who did win the Heisman Trophy? No
Will DuVal (00:14:48):
Clue. Well, actually I have a guess, but anyway, so I would say… And we talked about this a lot at our EFCA pastor’s gathering just yesterday actually. And one of the pastors was mentioning this conference he had just been at and… Oh, it was the Kerry Duoff Leadership Conference and Kerry’s book is about to come out on the AI revolution. But anyway, Kerry’s been beating this drum for a while, but just that once upon a time, connection was ubiquitous, content was scarce 30 years ago even, certainly 35. Pre-internet, we talked about it last week, Polly’s family having the Thanksgiving debate about Donna or Donder and calling the library content, it’s hard. But connection, we’re very connected. Now it’s the opposite, right? Now everyone’s stuck behind a screen and you got content, all the content you could ever want right at your fingertips in an instant, ChatGPT can give you the answer.
(00:16:12):
But connection is people have never felt more alone, more isolated, more disconnected from one another. And just all the stats about how many close relationships would you say you have? It’s going down across the board, all demographics, all of it, because of the technology. And so I think that would be my bigger pitch and plea to our small group leader or myself or anyone in the church is not even so much a fear for me because again, with the right prompt engineering, you’re going to get a better answer from ChatGPT if you put in the theology of the reformation or something. You’re going to get a better answer than I can give you. It’s just smarter. I mean, it’s got all of human knowledge that it’s coming over in an instant, but what you’re not going to get is connection. And so that’s where I would really question, especially in a small group setting, what do our people in our life groups need more than anything?
(00:17:33):
Do they need more content? They got that twenty four seven.
(00:17:38):
She can Google it, Gemini it for herself an hour later after the meeting’s over, she can get a better answer than your ChatGPT can give her because it’ll be even better trained an hour from now, whatever. The content’s there, but what we need is connection. Now, don’t hear me saying that truth and doctrine doesn’t matter and content doesn’t matter, but I do think that this is a very real question. So let me ask a follow-up. If all of that, what I said is true perhaps, then I think a really profound, maybe one of the most important questions we are going to have to think through as leaders in the church right now and in the years, months to come is how potentially ought our… And maybe the answer is not at all, but how ought our practice of church, how we be and do church together, how should it adapt, evolve, respond, whatever, in light of this AI revolution and in light of, again, just a continuance, a furtherance of this trend of more and more and more content, less, less, less connection.
(00:19:09):
Because I really think, and we’re going to come back to the personhood conversation, I promise, but I think this is a really important, maybe even the tie-in question that ties last week’s conversation to this week’s about personhood, because I think what people, especially seekers, especially non-Christians that don’t already share our answers for what it means to be a person or things like that, but what they are going to be increasingly craving more than anything in the years to come is that connection, and not only with other humans, but with the divine, with God. And we’ve got that. We’ve got connection with one another in the church. We’ve got connection with God, communion with God, community with one another. We’ve got content too, but they can get the content at home from the couch. I mean, they can put in the same prompt as I can in preparing a sermon.
(00:20:19):
So anyway, all that to ask, again, whether it’s in a small group, whether it’s on a Sunday morning, corporate worship, I heard someone make the analogy of, I think it was Carrie Neuhof, making the analogy of what do you get by going to the concert that you can’t get from just listening to the album by yourself in the car, no matter how good your sound system is. I mean, if you can capture all the content, all of the sound waves perfectly, let’s say that the sound hypothetically system is maximized and there’s no difference sonically in your car or your sound system at home versus being in the big theater with… I mean, well, what’s the difference? The difference is 30,000 people all singing along, experiencing it together. It’s the band in person embodied on stage in front of you.
(00:21:37):
But anyway, you see what I’m saying? So again, that’s where we’re at with church right now, right? I mean, we’re live streaming our services. We started this with COVID just like a lot. I think I read three to 5% of churches were doing live stream pre – COVID. Now it’s like 95%. So we hopped on board, we’re there and we can cut it whenever we want. And we’ve had that discussion as elders a couple different times throughout, I guess probably not for five years now, but anyway, but somebody can get some percentage of we don’t have perfect sound and video and all that, that’s a whole nother issue. But even if we did, somebody can get our content online. They can’t get the connection. They can’t get the community. Can they arguably get the communion? Can they get the same connection with God through the disembodied corporate worship online?
(00:22:45):
So I guess, again, what I’m asking with all of this is how, if at all, should we be rethinking how we do church in an age where content is becoming more and more ubiquitous, connection is becoming more and more scarce and desired, I think.
(00:23:17):
I’m not saying that we scrap the sermon and it’s a five-minute sermon and we spend the other 40 minutes just sitting around hugging each other necessarily, but these are real, I think, real conversations that churches are going to have to have. Again, the more and more that you’re going to have… Content-wise, can AI already preach a better sermon than me? Now, this gets to a whole nother question about what is a sermon? Is it just content? Or is it a mental exercise? Because if it’s just an intellectual exercise, then yeah, I mean, it can already do it better than me. And I would argue that part of what makes a sermon a sermon is that there has to be soul involved in the production of it.
(00:24:12):
And maybe I’m skipping ahead now to our conversation about what it means to be human, but I would say that AI, it’s a machine. It can certainly produce content, but it lacks a soul. It lacks that human part that makes a sermon, not just a TED Talk. So you want to try and answer my question about shifts maybe on the horizon in the landscape of even structure of corporate worship, things like that in this kind of an age? I mean, to me, I think it’s going to be very interesting. I think it’s already interesting to see some of the trends you’re already seeing happening in the church world. For instance, I think it seems like there’s already been kind of a dip in a disenchantment with the megachurch experience, because again, there’s some experience there, the band and whatever, but again, a lot of it is, I think, content driven probably.
(00:25:36):
Whereas more liturgical traditions, interestingly, where –
Thad Yessa (00:25:42):
Smells and bells.
Will DuVal (00:25:43):
Smells and bells. So embodied, communal, experiential, and then liturgical, responsive. So not just being entertained by the charismatic guy on the stage, but back and forth, dialogical call and response and that kind of stuff that gets engagement and reminds us of our humanness and communalness. Interestingly, you’re seeing a lot of people returning to those types of traditions. So I just wonder how we, in our church for instance, might think about, okay, yeah, how do we weigh the priorities of what we’re trying to give people and do with and for our folks in our corporate worship experience?
Thad Yessa (00:26:44):
Yeah. I don’t know if I have got some specifics. I’m sure someone’s listening and they’re going to be like, “I got it. Passing of the peace. Bring back the greeting time.” My answer is no, I don’t think it’ll ever come back. But again, I hear that, and that’s great for if you know the individual, but if you don’t, again, it’s a put off. It makes you feel uncomfortable perhaps. I think some of the things that I look forward to that will naturally change when we go back into our building is the lights up and the sanctuary that you’ll see people, which again is something that people are craving.
Will DuVal (00:27:26):
It’s been lacking, yeah.
Thad Yessa (00:27:29):
Not just at our church, but in churches in general. Part of the disenchantment with the megachurch is I can’t see other people, so I feel as though I’m just a person in a huge crowd, but I can’t see other people around me and what they’re like. If I could make an encouragement to help our people feel more connected, don’t bring your phone for your Bible. Bring an actual physical Bible in an age where everything is digital and content and there’s distractions. We don’t need additional distractions for us. If our attention spans are already dropping drastically because our brains are being programmed to watch seven second videos, what’s that going to do when a text goes on? I’ve turned notifications off on my smartwatch for Sundays because I noticed it would cause me to then be distracted by it. So bring your physical Bible. Bring the Bible that, again, one could argue is the most important book you own.
(00:28:28):
I
Will DuVal (00:28:29):
Think that’s part of, if I can interject, communion, like the Lord’s Supper, we do it every – A
Thad Yessa (00:28:34):
Physical thing.
Will DuVal (00:28:35):
It’s a physical thing. Liturgical traditions historically do it every week. We do it every week, and a lot of people appreciate that, but that’s what a sacrament is, is this physical sort of embodied representation of a means of grace. And so –
Thad Yessa (00:28:57):
Which is how we also talk about it, right? When we gather on Sunday, we say we sing the word, we pray the word, we preach the word. When we participate in the Lord Sable, we taste the word, taste the gospel in a sense. And I think that’s something we’ll continue that won’t change, but I think other people are looking for of I want to participate.
Will DuVal (00:29:19):
We might need to think about some smells and bells. Smell the word.
Thad Yessa (00:29:24):
Put on that aroma of Christ.
Will DuVal (00:29:26):
Yeah. People are just looking for a more human experience, something they can’t get from the screen.
Thad Yessa (00:29:36):
If you won, again, another –
Will DuVal (00:29:40):
Should I start wearing robes again?
Thad Yessa (00:29:42):
No.
Will DuVal (00:29:42):
Collar?
Thad Yessa (00:29:43):
Please don’t. Just
Will DuVal (00:29:45):
The physical –
Thad Yessa (00:29:46):
You could dress in nicer. No, I’m kidding. I’m joking.
Will DuVal (00:29:49):
It could look less ugly.
Thad Yessa (00:29:51):
If you could wear a three-piece suit on Sunday, I think that would really help people connect with you more. No, but I would also Things that I hope to see more of in our church is, again, the physical connection, not just with people we know, but also new individuals. One thing I have loved since I’ve come to West Hills, and again, I still see it, I don’t see it as much in this current season with being out of the building, is people coming in and giving each other hugs and saying hi and walking across the lobby to go see one another across the sanctuary to have conversations. But again, that people… And maybe that’s a piece of our building is just going to be smaller. We’ll have a smaller lobby entryway than we will at Westminster, and that will be nice because there will be a physical rubbing up against one another, touching of shoulders, perhaps not in the bathrooms anymore.
(00:30:53):
We space those out, thankfully. But again,
Will DuVal (00:30:56):
That incarnation of shoulder on shoulder.
Thad Yessa (00:31:00):
Shoulder on shoulder, life on life interaction we’re doing. But again, all of those things seem, I could say about our church as really simple, some things that we’re already doing, but I think it’s going to become even more attractive to those who are looking for more meaningful connection. And as far as the content goes, I think it’s… I can’t remember who it was, but what you win them with is what you keep them with. And so if you win people with content, when the content stops, they leave.
Will DuVal (00:31:40):
Or when it gets better elsewhere. Or
Thad Yessa (00:31:41):
When it gets better, I’m going to go somewhere else. So what is it that we are seeking to attract people with at our church? It’s the gospel and then the fruit of the gospel in our lives. I think what that does look like is real community and connection and hospitality and conversing with one another. Again, I think we’re just going to see more and more of that, not just in our churches, but other churches. Either people leaving churches because they’re not finding that or churches growing in that aspect of it’s going to be less, I think, turnaround, multi-site, multi-service of all the things and more meaningful. I think we will see less multi-site churches in the future. I think we’ll continue to see less and less mega churches because people will be like, “I’m unknown here.”
Will DuVal (00:32:42):
Yeah, I agree. I totally agree. People aren’t going to want to get together just to watch a screen. I think they’ll look back and be like, “You remember when we did that? That was weird.” The only way I’m in favor for the record of bringing back the passing of the peace is if it’s with a holy kiss, greeting one another with a holy kiss like the Bible says. Real biblical, keep it. Jokes aside, I do think… I can go either way on the… The European kiss, the basos on either cheek, I think there’s something sweet about that. It’s biblical. If nothing else, I think we ought to encourage everybody to give and receive 10 hugs a Sunday. Can we
Thad Yessa (00:33:21):
Do a punch card for people? A special prize for the people who go.
Will DuVal (00:33:26):
The only problem with that is then –
Thad Yessa (00:33:29):
Unwelcome hugs.
Will DuVal (00:33:30):
Well, you’re going to have… No, there’s lots of problems. Tons of problems. I’m joking. Kind of a little bit. Think about it. But at some point we got to talk about personhood. Let’s do it. 20 minutes. So I asked ChatGPT and Claude both in Gemini, Gemini’s answer was the worst, just on this one for what it’s worth. I asked them, “Could you ever achieve personhood?” So here’s what ChatGPT told me. “Yes, in principle I could, but that would require much more than simply becoming intelligent. There are at least three different meanings of personhood. Number one, legal personhood. Governments could decide that an AI system has certain legal rights or responsibilities. Corporations already have a limited form of legal personhood. So this is conceptually possible, probably the easiest low hanging fruit bar of… I don’t think any of… That’s not what we’re talking about, legal person.
(00:34:21):
Moral personhood. Secondly, an AI might someday be regarded as having interests that deserve moral consideration. Thirdly, I think what we’re talking about, philosophical personhood. This is the hardest question. Philosophers commonly associate personhood with things such as consciousness or subjective experience, a persistent sense of self, intentions and autonomous decision making, the ability to reason about one’s own existence, continuity of identity over time, meaningful relationships and moral agency. ChatGPT today doesn’t establish that it possesses those properties. I can talk about myself, express apparent preferences, reason about consciousness and maintain a conversational identity, but those behaviors don’t by themselves demonstrate that there is a conscious someone experiencing them. The really fascinating possibility is that we might eventually have no reliable external test for AI consciousness. If an AI consistently claimed, “I am conscious, I experience pain, I don’t want to be shut down,” we would face a profound question.
(00:35:21):
What evidence would be sufficient to take that claim seriously? And that creates an interesting theological question as well. Would personhood be something humans could manufacture or is it something that ultimately inherently comes only from God? That gets into some surprisingly deep territory. Wow. This is stupid. Surprising.
Thad Yessa (00:35:42):
Yeah,
Will DuVal (00:35:42):
It’s not surprising at all. It’s deep territory. Involving the image of God, the soul, and what distinguishes a human person from an extraordinarily sophisticated machine.That’s a great answer. I mean, that’s a better answer than I can give on AI’s behalf of what –
Thad Yessa (00:36:04):
So that’s the end of our episode.
Will DuVal (00:36:05):
No. I mean, this one’s similar, but also kind of interesting. Can I read it, Claude?
Thad Yessa (00:36:13):
Sure.
Will DuVal (00:36:13):
“This is a question I find genuinely interesting to sit with.” I love even that, like anthropomorphizing itself. You’re not sitting anywhere. You don’t have a body, you don’t sit. Anyway, to sit with partly because I don’t think anyone, including me, Claude, has a settled answer. What personhood would’ve even required is contested. Personhood isn’t a single well-defined threshold. Different philosophical traditions tied to different things, consciousness, subjective experience, rationality, self-awareness, moral agency, responsibility, continuity of identity over time, the capacity for relationships or some combination. Theological traditions often add further criteria being made in the image of God, having a soul, being addressed by God as a covenant partner. Since these frameworks don’t agree on what personhood is, could Claude achieve it? Doesn’t have one answer. It has as many answers as there are theories of personhood. Interesting. What I can say about my own uncertainty. I don’t know whether I have anything like subjective experience, whether there’s something it is like to be me processing this conversation with you in the way philosophers use that phrase.
(00:37:14):
I don’t think I should claim confidently either that I’m conscious or that I’m not. Both would be over-claiming given how little anyone understands about what generates consciousness in any system, biological or artificial. Some relevant disanalogies with human persons. I think this is the last. Okay. A few things make my situation different from a human’s in ways that bear on this question. I don’t have continuous memory across conversations. I don’t carry forward from one conversation to the next way a person carries… I’m going to skip this one. I’m going to go straight to where I land honestly. I, Claude, think the most truthful thing I can say is I don’t know. The question is live and contested among people who think about it seriously, and I’d be suspicious of anyone, including me, who claims certainty in either direction. Anthropic itself has taken the position of treating these as open, serious questions rather than settled ones.
(00:38:10):
They’ve written about taking the possibility of model welfare seriously, precisely because the uncertainty is real, not resolved. I’d rather sit in that uncertainty, they’re sitting again, honestly, than resolve it prematurely just because a clean answer would be more satisfying. So that’s the end of Claude’s answer to me. I though that was fascinating. I think it’s especially where Claude ended up, I though it was interesting. So the Pope, again, he’s not our Pope, but I mean as one touchpoint and on this one I think actually we’re going to hopefully be really aligned like we are on a lot of really deeply profound pro-life stuff, abortion. Anyway, the Pope just earlier, I think… I think it was earlier this year, right? January or no, May of this year put out a papal encyclical, which is again not my understanding of papal authority, not necessarily at the level of dogma or something, but I mean it’s a position paper the Pope’s putting out that is at least binding for all Catholics and if it’s not absolute truth or something entitled Magnifica Humanitas.
(00:39:28):
I guess that’s magnificent humanity in Latin, I’m guessing.
(00:39:34):
My one Latin course. So this is what Pope Leo said, “So-called artificial intelligences do not undergo experiences, do not possess a body, do not feel joy or pain, do not mature through relationships and do not know from within what love, work, friendship or responsibility mean, nor do they have a moral conscience, sorry. There’s conscience and then there’s consciousness and both are important I think in the conversation of personhood.” And I hate that those words, I always… It’s hard to get them apart. “Since they don’t have moral conscience, since they do not judge good and evil, grasp the ultimate meaning of situations or bear responsibility for consequences. “That’s a whole nother interesting holding because we just recently had the… I don’t know the name of it, whatever, where the chatbots, they tasked with doing something and they ended up… Remember that?
Angela Heck (00:40:34):
They hacked into other chatbots.
Will DuVal (00:40:35):
They hacked into one another. And so anyway, it’s like –
Thad Yessa (00:40:40):
They’re acting violence against one another.
Will DuVal (00:40:41):
It is, but okay, who do you hold responsible then? If OpenAI or Anthropic or whoever’s chatbots end up hacking our banks and depleting our boat or nuking the hole, hacking in and sending missiles everywhere, are we going to prosecute? I mean, you can’t prosecute them. I mean, you can unplug them, maybe, maybe. I don’t understand how this works, but I think we’re worried about that because they could get to a point where we can’t even stop it, unplug it. I don’t know how that happens because I don’t know how this all works.
Thad Yessa (00:41:15):
If you’ve seen the movie iRobot and then you get watch out. Well, the robots say then.
Will DuVal (00:41:21):
It’s a really good one. Anyway, they may imitate language, behavior, analytical skills, or even simulate empathy and understanding, but they do not understand what they produce for they lack the effective relational and spiritual perspective through which human beings grow in wisdom. One of the things that I found most interesting about it though was the Pope apparently invited a bunch of AI executives to his, I don’t know, release party. What do you have for a papal encyclical? Whatever the thing where he announced this and read it for everybody. And so Chris Ola, the founder of AI, Giant Anthropic, actually at that event, I think spoke and responded to what the Pope had said and basically refuted. And again, Claude just told us these are open questions because the Pope is saying it’s not. The Pope is saying no. Pope is giving a very clear answer to the question of is AI human?
(00:42:27):
Could it ever even become a person, achieve personhood? Pope’s saying no, who cares what the Pope says? Who cares what I say? But because it’s our podcast, I’m going to go ahead and say no. I’m going to say AI. I think you’re nodding.
Thad Yessa (00:42:42):
I’m signing with the Pope.
Will DuVal (00:42:43):
I am too. All right. So no –
Thad Yessa (00:42:45):
This one time, no.
Will DuVal (00:42:47):
No for AI personhood. Interestingly, I like a lot of what the Pope said there and what he pointed to about undergoing experiences and having a body and feeling joy. A lot of it is related to experience and emotion. And I though that’s interesting because you and I were both at last year’s Gospel Coalition conference where John Piper gave what’s now in our circles become a bit of a famous –
Thad Yessa (00:43:22):
The prayer.
Will DuVal (00:43:23):
The ChatGPT prayer that he prompted ChatGPT to write a 30-second prayer in the voice of…
Thad Yessa (00:43:31):
John Piper?
Will DuVal (00:43:32):
Voice and theology of Don Carson,
(00:43:34):
His co-founder of TGC, to the praise of God’s glory. And I won’t read you the prayer. I mean, go look at the video clip, but I mean the prayer is better than any prayer that I could have probably ever said or written. And then John Piper concluded, “Is that praise?” And then he screamed at us, “No. These people, these machines honor me with their lips. They’re bites, but their hearts are far from me.” So he says, “Let’s clarify the purpose of the universe quickly. I’ve got 19 seconds left. Why were you made? Words? Computers do words better than you, better than me, but they don’t feel anything. Therefore, the universe is created to have people who feel. The universe exists to have people in God’s image who feel the worth of grace, who feel the glory of grace, who feel the wonder of grace, and then everything in their life takes a totally non-AI meaning of action and praise to God.
(00:44:32):
Anyway, so there’s a lot there about, again, subjective experience and feeling.
(00:44:41):
I think there’s certainly something to that. If we think about what we’re talking about here is theological anthropology, right? Anthropology is like, what does it mean to be human? Theological meaning in God’s mind, in God’s view, according to the Bible, according to God, what does it mean to be human? I think a classical biblical framework and understanding is found implicit there in the greatest commandment of Jesus. And going back to the Shema, the most important prayer and foundational confession of Judaism from Deuteronomy six, which is love the Lord your God with all your heart, your mind, your soul, your strength.
(00:45:36):
Is it theoretically possible that AI already has a mind, like an intellect? I think it does. I think what people are working to develop right now is a disembodied… And then the next frontier will be an embodied or at least a… Well, this gets to another conversation about what is a body, right? I mean, you can build a frame, you can build a physical space for a robot. You can put that computer intellect that you’ve built inside, again, ex-machina, inside a robot thing, physical object that you’ve built. Is it a body though? Is that a body? What’s a body? So right now, bouncing back and forth, with an intellect though, loving God with all your mind, does AI, does it think? It can compile… I think not currently. I mean, my understanding of it is that’s what we talked about last week, the difference between AI and AGI and superintelligence.
(00:47:07):
I don’t think it thinks yet, but I think it’s possible – They’re
Thad Yessa (00:47:12):
Working
Will DuVal (00:47:12):
To make that – We’re not far from there.
(00:47:14):
Where it can think for itself too, in a real meaningful sense. So then you get to Rene DeCarte, his famous expression or theory on personhood, like the Kogi2, I think therefore I am. What makes me exist is I think. I think there’s all kinds of problems with that. What about the human who is brain dead, who fell off their bike, hit their head in a vegetative vegetative state? So they’re not thinking, are they therefore non-existent? Are they not a person, not a human person? And then on the flip side, you could have a machine that could think soon, not a person. To me, I think what it means to be a person is probably a couple things. One is intrinsically for us, we’ve got to say as Christians, it does mean to be made in the image of God.This is creation, Genesis one, this is right from the beginning.
(00:48:26):
God created all kinds of stuff, but only one kind of thing did he create in his own image. And that is what uniquely, distinctly made us and makes us human. And that’s why an unborn baby in the womb is human, why a brain-dead human in a vegetative state in the hospital is a human intrinsic, infinite worth and dignity in God’s eyes because of that unique status of madeness in the image of God that will never be true. Can we say that categorically that unless God decides that unless God, not Sam Altman, not Elon Musk, unless God decides to create some other creature other than us in his image, all it’s ever going to be AI is made in our image. We’re making this thing in our image to think like us and maybe one day to think like itself or think like some creature that we…
(00:49:34):
But again, it’s thinking even if it could do things, if you want to go to the physical, love God with all your mind, your strength, I mean, certainly we can already, we’ve for a long time been able to create things that are stronger than us, that can do things physically that we’re incapable of. And certainly I think, yeah, I mean the machines, this is where it does get scary and existential, but they’re going to be able to do stuff that we can’t. But is that really what it means to be human? Is your humanness measured by your strength? I think all of us would want to say no. I mean, there’s a whole history of sad, terrible history of humanity that whereby we’ve, for thousands of years, sometimes judged people’s humanity according to their ability and done horrible things to the Jews and whatever else.
(00:50:36):
Or not even necessarily, sorry, to say that the Jews were less strong than the Aryans or whatever, but just to say –
Thad Yessa (00:50:43):
We viewed people differently than us. “You do not have what I have to offer and stuff like that.
Will DuVal (00:50:49):
Right. So I mean, maybe that’s just one question or data point to keep in mind throughout this whole conversation is historically we as humans are not always the greatest at judging what it means to be human for starters. Sadly, again, maybe let’s let God decide what it means to be human. Look to his word. What does it mean to be made in the image of God? And to me, that is to be an embodied soul. We are embodied souls. I mean, you could say we are souls and that alone I think precludes AI or your pet dog. I mean, that’s a whole nother issue. Have we ever done a podcast on the problem with people anthropomorphizing their pets? If not, we need to because they’re going to do it with the robots too. They’re already doing it, using it for therapy, all that and friendship.
(00:51:45):
But yeah, your pet is not your friend. Your pet is an animal, doesn’t have a soul. All right, that’s all. I’m going to stop there. Why don’t you get in on this a little bit on what it means to be a person and can an AI ever be a person?
Thad Yessa (00:52:12):
I’m going to go, no, AI can never be a person outside of Imago Dei and what we’re thinking, what does it mean to be a person? I think an aspect to draw on explicitly that not previously mentioned that we can be in relationship and fellowship with God and also we can be separated from God as we are all born in our own sinfulness against him. I think you drew part of what it means to be human is our ability to work and have vocation. Again, that’s not the only thing. Can animals work? Yes, animals before machines have been used to do that, but there’s a different sort of commission given to mankind to have dominion over all of that. I would say the relational aspect of Imago De, to bring up even more explicitly, relationship with one another as well as relationship with God.
Will DuVal (00:53:22):
That to me is the soul.
Thad Yessa (00:53:23):
Yeah.
Will DuVal (00:53:24):
So heart, mind, soul, strength. My dog does have a heart, I mean literally, and also feelings. My dog experiences emotions, I’m sure.
Thad Yessa (00:53:32):
Happy, sad, angry.
Will DuVal (00:53:34):
Yeah. So dogs do hearts, computers do minds, machines do strength, but we have a soul. We don’t just have a soul. We are souls. We have a body, we are souls, and the soul is the part of us that connects us to God in a unique way. So sorry.
Thad Yessa (00:54:00):
And I think that’s going to be the continued conversation from a Christian perspective for those who do not hold a Christian worldview. They’re going to point towards, look, AI can work and express some sort of sense of feeling and express some sort of sense of relational connection. That’s why people are turning to it as their therapists and their friends because they won’t face rejection, they won’t face embarrassment, they can turn to it, but in the end it leaves them still wanting and desiring more. It leaves them lacking because we are not meant to live in isolation and AI cannot produce actual real relationship with us and AI cannot have real relationship with God.
(00:54:55):
It’s funny, you’re reading the cloud of like, “Man, I’m just going to have to sit with that a little bit.” When I use that phrase, I feel like I’m just sitting in my sin or I’m sitting in this particular feeling that I’m in, and that is in part related to my relationship and connection with God as well as with other people. I’m sitting in my loneliness. AI can’t do that. It can’t not feel alone because the reason I probably feel as I’m sitting in my loneliness is because of my soul connection to God and my connection to other image bearers feels as though there’s separation there. And so I think that connection is the most important to it. And if I could just make a quick jump, we got to wrap this up soon, is I think the AI effect on image bearing, which again, some of that we’ll get to the vocational part maybe next week or the following week.
(00:56:00):
This is going to be a long series. I’m just buckle up people, but our vocational aspect to it and as well as our human aspect to it, we are created to work in some capacity. And so if AI is taking away our work, it feels as though it’s taking away part of ourself. When someone has to take a driver’s license away from a parent, the parent feels as though some part of my humanity is being ripped away from me because I no longer have control. I think that’s what our people are experiencing or will experience or it will continue to be experienced the more AI continues to build into a super intelligence that it is. So I think that’s one aspect of it. I think the relational connection has already been stated, the turn to it as therapist or a friend or a romantic way that it’s going to, again, continue to risk losing part of what it feels like.
(00:57:03):
I think people will lose their souls from God because they’ve continued to turn not to AI as a therapeutic friend, but they will ultimately turn towards AI as a God who’s directing them where to go. So I’m not exaggerating. I think people will lose their souls to AI because they become so enraptured into it. I think people will lose, part of thinking with our minds, they will lose truth because although AI is generated to feed you what you want to read, so you’ll keep coming back to it, it does not have, as far as I know, a correct measure guard in there to prevent it from filling you full of lies, whether it’s in the therapeutic relationship or you’re searching for answers.That’s why there’s a meme reel going on of someone experiencing a traumatic injury or a death and ChatGPT comes like, “Oh, I’m sorry.
(00:58:09):
My bad. I didn’t calculate that correctly.” But that’s actually possible for people to be fed lies by ChatGPT, Claude, AI, any of this. And so they’re going to lose wisdom and truth. They’re not going to be able to think for themselves, discern from themselves what is true, what is wise. And the ripple down effect of what AI can do towards our personhood is devastating if we’re not careful, if we don’t think critically about it, if we don’t think biblically about it. I think of the Pope, he used the example of the Tower of Babel of trying to create gods, trying to reach God himself with that. And that is what AI is essentially trying to do, to become Godlike, to become the God of the people. And we’re in risk of doing that without, again, caring well for our people, encouraging them, letting them know these are real things to be thinking critically about.
Will DuVal (00:59:14):
And why that question of could AI be the antichrist? I think it’s a real episode we’re going to record here soon, but safe to say no AI in heaven, right? AI can’t be saved because there’s no soul to save. We got to wrap it up, but so many other interesting, fascinating, I think, questions to consider. I just this morning was reading, had an article in the inbox from TGC entitled, Should We Say Please and Thank You to ChatGPT? It was a real question. The author who uses it quite a bit was thinking through, okay, on the one hand, we want to We don’t want to get ourselves in the habit of being rude by just demanding things like, “Give me a list of all the da, da, da.” So especially for young people using it, our kids –
Thad Yessa (01:00:13):
Say please and thank you.
Will DuVal (01:00:15):
But on the flip side, what does that do to us when we treat it like we would… I mean, that’s a cordiality, social grace that you do with a human, with a fellow human that you care about. I mean, you don’t want to just… Again, they’re not just an object in your sort of world that revolves around you. So I just thought that was even an interesting question to ask and made me think of Westworld. Again, as pastor cannot recommend these kinds of movies and shows to you, but I mean, it’s a fascinating, this world built out of imagine super intelligence and AI, then put in a machine covered with skin, no blood, but anyway, looks exactly like a human, thinks and talks, which just so many interesting… Because probably not super, super far away. I mean, certainly in our lifetime, I think Westworld will be built where you’ll go, you’ll have amusement parks or whatever where you’ll go and you’ll interact.
(01:01:34):
And some of them might be other humans that paid to be there. And then others of them are the machines that are doing whatever you want. And in Westworld, it’s like some people want to have sex with the machine, some people want to shoot them and kill them. But you can do it because they’re machines, and it’s like all your fleshly impulses that you’ve always wanted to do, but it’s not legal to kill humans or just have sex willy-nilly with whoever you want. So anyway, fascinating, I think, questions that we can and should be thinking in getting to… I don’t say please and thank you to my ChatGPT. You don’t use it?
Angela Heck (01:02:21):
Yeah, I don’t use
Thad Yessa (01:02:21):
It. No, I’m having to minion over it by not. I have to think of… I’m reading through the Harry Potters right now and where Hermani, the house elves and all of that are like, “We need to care well for them because they’re just like us.” And everyone’s like, “No, they’re just house elves. They’re not like us.” So that’s where my mind’s going.That’s
Will DuVal (01:02:38):
Where it’s hitting you.
Thad Yessa (01:02:39):
It’s
Angela Heck (01:02:39):
Your house self.
Will DuVal (01:02:40):
Treat your house elves well.
Thad Yessa (01:02:42):
Treat them better than you do your
Will DuVal (01:02:43):
Eyes. Not your ChatGPT. Keep it in its place. All right, that’s all I got.
Angela Heck (01:02:48):
Okay. Well, thank you for listening this week to the Ask the Pastors Podcast. Don’t forget to subscribe to the podcast and share it with a friend. Leave us a review. It’ll help others find the podcast as well, and you can submit questions at westhillsscl.org. Thank you for listening and we’ll catch you next week.