Digital Deacons: AI
252 | “Gone Off His Rocker”
Put AI to Work for the Kingdom
Grab your nerd glasses — our monthlong Geektober deep-dive is officially here! First up we head to the workshop to explore Artificial Intelligence. Is AI a faithful "digital servant" for your Bible study, or a rogue robot that hallucinates answers? We're giving you the guardrails to use this powerful tool wisely, showing you why it's a great research assistant but a terrible rabbi. Join us for a fun, nerdy, and practical guide to making AI your servant, not your master, in your walk with God!
Takeaways
The Big Idea: Artificial intelligence is a powerful tool that Christians can leverage for kingdom work, but it should be treated like a helpful servant, not a replacement for critical thinking.
This Week's Challenge: This week, put an AI to work as your Digital servant. Then, evaluate the result. Ask yourself: "How did this information help me to think more deeply and personally about the text?"
-
Introduction and Welcome
Bryan: People are going to know right off the bat like Ryan has gone off his rocker. Well, hello everyone and welcome to the Bible Geeks podcast. This is episode 252. I'm Bryan Schiele
Ryan: I'm Ryan Joy.
Bryan: and thanks so much everyone for tuning in, you know back in the day Ryan and I served as the original digital Deacons using the best tech that we had to serve the church that we were at and now technology has taken a quantum leap forward with
Geektober Kickoff: Embracing AI
Ryan: Today in our Geektober kickoff, we're exploring how to responsibly enlist AI as your own digital servant, a powerful assistant for Bible study and ministry without letting it become a replacement for your own heart, mind, and discernment.
Bryan: ready to go.
Ryan: Let's get into AI.
Bryan: When was it that you put into our base camp a notice about some weird thing you had in mind for October? 'Cause that was a moment when I was going through our planning that I was just immediately, like, unlocked in terms of all nerdiness that could be possible. What was your idea for this "Geektober" as you called it back then?
Ryan: Well, I was just thinking, we call ourselves the Bible geeks, and we can geek out with the best of them, but what if we really unleashed our full geekiness
Bryan: [laughter]
Ryan: and allowed ourselves to get into a little bit, it's almost like a license to get a little bit more, a little deeper into some interesting stuff that we're thinking about or we think would be fun to talk about that really, we maybe wouldn't preach on it as often as say, joy or faith or obedience or something like that.
Like for instance, dragons and giants and AI.
The Role of AI in Ministry
Bryan: Yeah, if your really bad dad joke at the end of the last episode was any indication, we're in for an interesting conversation today where we talk about AI in our work in the kingdom, but I do think it's important for us to kind of roll back to this idea of Geektober. over the past, I don't know, how many years we've been doing the show, it feels like we've done a lot of, like, really nerdy, geeky kinds of things, and we've leaned into that a lot, but then you wonder, "Could we be any more nerdy?"
And I guess we'll find out here in this upcoming Geektober month that we're in right now. I'm excited about this conversation about AI. This is definitely in my wheelhouse. Is this something that you've been noticing more in your experience around asking people, like, hearing people's discussions out in the world, abroad?
Ryan: yeah, there is, it is in the zeitgeist, as they say. It is all over. Everybody's thinking about it. A lot of people I talk to are afraid,
Bryan: Oh, yes.
Ryan: but there's conversations all over. People are excited about it. I see it showing up in like texts that people send me. Hey, here's some ideas for, a class we could do. And I just look at the formatting and I know, oh, okay. They did some work with ChatGPT and sent that over.
So, everybody's kind of experimenting and learning as we do when a new technology shows up. And everybody's wondering, hey, is this gonna be dramatic in the way the world is changing, or is this gonna be just another step in our constant shifting world,
Digital Deacons and Technology in the Church
Bryan: You talk about technology, and for those who know us and who know our backstory, you could also go on to biblegeek.fm/about and find out a little bit more about this. But Ryan and I used to be digital deacons at the Monte Vista congregation where I'm currently attending, and that was really an idea focused mostly on our involvement with the website and anything that had an internet connection.
So, like, even our thermostats at the building were all Wi-Fi connected, and all the computers we used, and the recording equipment, and the AV system, and everything else. That was within our purview. And so you and I worked pretty well together in getting all that stuff up and running and really just pushing that forward.
Ryan: I feel like more than the website, the really new thing for elders to contemplate that we brought to Monte Vista was social media.
Bryan: Oh, yeah.
Should a church
Ryan: be on Facebook? Is that too social? Is it dangerous somehow? that was a moment where people were not sure what to do with social media.
And you look now, and it has changed a lot of things in some bad ways, some good ways, but it was worth thinking about, and responsible people ask good questions about any new technology as we determine how to wisely use it in the Lord. And that's the thing that I think people were trying to figure out.
We were working with it and teaching people how to use it in my company with our clients. And then it was like, okay, what conversations do we need to have with elders? So I don't know, what do you remember about that?
Bryan: yeah, social media was definitely an exciting yet frightening endeavor at the time. I think that was one of them. Tangentially related for me was video recording. So that was very new. And we were really putting out there all of our sermon videos, and, very tightly clipped so that they were just the sermon.
We were trying to do things like flipping back and forth between the slides so you got a better experience with it. And it started out really janky at the time. I think we had some, like, questionable webcam, like, mounted on the front pew or something. You had basically a
Ryan: Oh,
Bryan: up the speaker's nose, which was, yeah, that was solid.
but over the years we sort of worked on improving that. And that was something I was really excited about because that really was something that I didn't see in a lot of other congregations on their websites or on YouTube. And so we really, like, established a presence in recording our sermons and putting them out there.
you know, they say that content is king and, definitely that was the case, putting up all of this video and all these resources available on our website and on YouTube. And it was a questionable, like, "Is this good? Is this bad? Are we sure about this?" And it turned out to be, yeah, and now, all
Ryan: know, post-COVID, everybody has videos of everything. But I mean, well, that's another thing. I mean, that wasn't when we were digital deacons, but whenever we were, all churches had to navigate something very new and strange during that COVID process. And in some ways, the church has been blessed by it, as a lot of people updated their technological capabilities for video, as you're saying, and other things.
But yeah, I think that stability and confident, I don't wanna say predictability, but I think stability is the right word, is really important in our faith, in religion. wanna know, hey, this has been around for 2,000 years. The Bible doesn't change, and so I can be confident what my, what my worship experience is gonna be, and all of those kinds of things, the things we come to rely on.
And so then, whenever you start introducing something new into it, we have, rightly, a lot of questions, and there's some uncertainty, a lot of sorting things out. And then you figure it out, and you move into the new era, and say, hey, we're gonna stay true to these ancient principles, but this update doesn't affect that, Yeah, exactly. It's a lot of the conversations that probably swirled around about the advent of, the overhead projector system and the, Yeah,
Bryan: presentations. Oh, yes. S
The Workshop: AI as a Tool for Bible Study
Bryan: o let's get into an actual conversation here about artificial intelligence in a segment that we like to call "The Workshop."
Oh, Norm Abrams. I love that clip so much. So we're really, in this episode, talking about AI in terms of being your "digital deacon."
And I don't think necessarily a tool like AI is super far off from that kind of role of a servant. Although, definitely not the same. Of course, you know, you're not going to see in the qualifications of a deacon anything that pertains to AI, I don't think.
Ryan: Yeah, I mean, I see where you're going with it. I don't love the comparison myself, because,
Bryan: good.
Ryan: what you're saying is servant, right? Diakonos means servant,
Bryan:
Ryan: I think the way we wanna characterize this is
the way Norm Abrams, is that who that was,
Bryan: Yeah. Oh,
Ryan: you know, this
is a tool.
So if I had to put it in one sentence, it's, AI is not a person, it's a powerful tool. And so we wanted to find AI, and we wanted to find person, which is really important, and I think helps us, from a Christian perspective, answer a lot of these questions. And then we wanted to find what a powerful tool this is, but, obviously, a deacon is given, whenever those proto-deacons in Act Six were put into place, they were chosen because they were men who were full of wisdom and the Holy Spirit.
And I bring this up, not to riff so much on the digital deacon idea, but to start this conversation about AI, and what this is and what it's not, and so a deacon, actually, is someone who is skillful as an image-bearer of God, led by God, to make discerning, difficult choices that only a human can.
And so that's what a person is, and that's what a deacon is. You know, it's not so much about, like, First Timothy Three, you know, is AI, is AI the husband of one wife, you know, that kind of thing?
Bryan: He's definitely addicted to much wine, for sure. There are times where he hallucinates pretty bad.
Ryan: But, no, I think that's a good kind of, like, side thing. But the idea of using this as a servant, it can be a really good, powerful tool, and we wanna be as fruitful as we can be. So with that little aside, maybe we can go into some of the ways that it can be useful in our workshop
Bryan: Yeah, and I think that is helpful. I will just go back to, it being a tool and I will vacillate between referring to AI as a he or a they or even a she and in a little bit here, but also as an it because it is an it. Really, it is a tool. It is effectively like a shovel or in Norm Abrams parlance.
It is like the table saw. It's a tool that we as humans wield for our advantage.
we all know the value and the power of capable tools. And so I think where we're headed here in this first part is really to look at some of the opportunities that a tool like AI can bring to the table. And I think for me, there's been a lot of areas that I've seen the power of AI really pop up.
But before we get too far down the line into like what it can do, you kind of alluded to it.
Understanding AI: Capabilities and Limitations
Bryan: What is it? From your perspective, Ryan, give me like the give me the sense that you get for what AI is and like what it's doing.
Ryan: Well, just, and you are more capable of answering technical questions about this as a software engineer.
Bryan: I want to know
Ryan: okay.
Bryan: answer.
Ryan: So it's really helpful to me to know that the language model apps that you would work with, like a ChatGPT, are probability engines that are just predicting what is the most likely next word.
It is not, it's not sentient, it's not thinking in the way consciously that we would, it doesn't know you exist, but it is really good at taking tons and tons and tons of data and synthesizing it in human language that is understandable, and it can understand my human language really pretty well.
Bryan: I think if you ever watch like when you were sick as a kid and you had to stay home from school and you got to watch The Price is Right, you ever watch The Price is Right on
Ryan: Oh yeah,
Bryan: So yeah, that's it, right? Plinko is actually the perfect analogy for what AI is doing.
If you think about what you provide to the chat engine or whatever, what you give it, you're basically putting in one of those Plinko, I don't know what those things are, like the disks or whatever. And you put it at the top of the whole board full of like pins. And as it goes down, it's weighted to be able to know, okay, when I hit this pin, I'm going to go left.
When I hit that pin, I'm going to go right. And it generally will be able to predict its way down to a reasonable response. And the way that these things are created is basically playing like a bazillion games of Plinko and then really landing on what weights along the way, what predictions along the way are needed to be made to get to a reasonable answer at the end of it.
And so that's a helpful way of thinking about it. Like every time you interact with it, it's using everything that you've previously provided it, all of the knows about you in everything that you've shared with it. And it's using that as a weight and a prediction to give you a response at the end of it, like a Plinko game.
And so,
let me give you an example of something that I have used.
Google has a tool called Notebook LM. Notebook LM is basically a place where you can take documents like PDFs or transcripts of things and you feed it into a database. And then basically you can ask the database for all kinds of questions regarding that information.
And it knows based on your questions what it's looking for in those documents and can synthesize everything together in a way that gives you reasonable answers and it cites all of the sources that you've provided it. So I actually have an entire Notebook LM for this show. So I have all of our transcripts that we've ever done for this show and I can ask the AI assistant.
I can say, hey, you know, when was the last time we talked about fill in the blank topic or, you know, what verses have we landed on in a Jesus said segment more than anything else? And it can go through and it can find the answers to those things and provide them.
So like as a as a glorified Google, like really that's kind of in a sense what these tools are really good at as a glorified Google. These things are actually very helpful in doing research even in like religious kinds of
Ryan: it's funny how we just take for granted that Google is, there's no problem with Google. Like the search results you get from a Google machine are, if you ask it a theological question, you know, what do I need to do to be saved? Do you feel like you can just absolutely rely on the first result or the second result or the third result that comes?
you have to use discernment, but it doesn't stop you from using Google. And so that's an interesting connection. It's a glorified Google. It is much more than that, and it does different things, but I think that's helpful.
Yeah, I bring up Notebook LM because it's kind of one of the simplest and most tractable services to understand in terms of AI. But yeah,
Bryan: I've actually thought about if this was something I was still like super engaged in at Monte Vista, I might have actually pushed to like have all the transcripts from all sermons pushed into a congregational wide notebook LM so that like our members could go in and say, hey, when did we study about this particular topic?
And then they could go back to and listen to that episode or like, you know, they could ask questions of, sermons that had been preached at the congregation years and years and years ago. But again, like Google, you're harvesting all that information, pulling it out and making it useful again. So a service like notebook LM as like a search engine or a research tool is very helpful in that regard.
Ryan: Okay, I have a question.
Ethical Considerations and Guardrails
Ryan: Do you wanna get into ethical things right now or like guardrails, or you wanna keep going? 'Cause what you just brought up was so interesting and makes me think, like the church application, but makes me think about like all the things that like the giant notice I would wanna have at the top of something like that.
This is not the Bible. This is not what our congregation believes. We believe the Bible, but here are some answers from thoughtful people who have thought here. You know what I mean?
Bryan: Yeah, that's a really helpful way of framing it, because as with anything you're dealing with the fallibility of your own preachers. Number one. Okay. And and this is why I like if I was going to have a tool like this available, I would make it very clear like, hey, this this is relying on a tool that may not be reliable and it's also relying on men who may not be reliable. So there's that aspect of it. I definitely get, you are any time and I think it's helpful for us to bring this up now in this conversation.
Anytime you are using a tool like this, you are asking for it to make big mistakes. And so you need to be aware of that. Number one, like, yes, these things are cool. They're good research tools. And like you already said with Google, I know when I go look at Google for anything, there's going to be some wrong stuff in there.
There's going to be people spouting off blog posts with bad ideas. There's going to be like Reddit threads that have no, value whatsoever. But then there's also going to be those nuggets of insight that I'm like looking for. That's what I'm searching for. And so, yeah, I think anytime you use a tool like this or make anything like this available to other people, big asterisk.
And it's why I think on the bottoms of most of these chat bots like ChatGPT, Gemini, Notebook LM, they almost always at the bottom say something like these tools will make mistakes. Please make sure you cross reference them. That's an important thing to do here.
Ryan: Yeah, there's so many, boy, I have so many questions for you and so many things to think through. Like, so you brought up the first thing you're gonna deal with is the fallibility of humans or of the data, whatever it is, the data that you're giving it. So if you make it a little bit better, maybe a lot better, whenever you say only search these things that have been taught here by people who hold our values, that the inerrancy of the Bible and these kinds of things, so you're limiting that more than just all of the internets of the whole world,
Bryan: Oh, yeah, I would never do that. That would be a terrible idea. So for me personally, I have a band of brothers Notebook LM, which is basically your transcribed Goodwin's, and a couple of other friends of mine that I trust. I would not sprinkle in just Joe random preacher off the streets information in there unless I had, had strong trust in that person.
Ryan: Yeah, yeah, so then there's the second level of that besides human fallibility, even with, as much as we love all of those people, there's also you brought up this term hallucinate, hallucinations, what does that mean when an AI hallucinates? Such a strange term for it.
Bryan: Yeah, so you can really start to see the cracks in the machine when you ask it for something and you can see that it does something that kind of makes sense, but is totally off the mark and with these models improving over the years. Hallucinations have kind of gone down quite a bit. In the beginning, you used to ask a question just a most, the most basic fact about something and it would go off and just make something up.
Like it's not giving you facts. It's not a fax engine. And so it's like it's finding something that in the probabilities will come up with a reasonable answer that it thinks is reasonable. But sometimes what it tells you is just like a bad dream.
let's let's go back in the image generation version of what we did a couple episodes ago and I showed you the vision of Ezekiel's throne room scene and all those things that were going on with the glory of God departing the temple and not even remotely close not even remotely close.
And so that's like a visual representation of hallucination. Like it's just making stuff up. So you really do have to check those things and that's why I keep going back to like you cannot trust these tools. However, I cannot trust that my table saw isn't going to you know, suddenly lock up one day and stop functioning or I can't I can't trust like you were talking about that Google is going to just give me results that mean nothing to me at the time.
It is again our wisdom we bring to the table here that has to be at play.
Ryan: if you were to make that notebook LM database and somebody asked it a question, you would just want them to distinguish from asking one of our elders a question, like me and our elders are doing a Q&A this Sunday night where we're just gonna go up there, we've asked people for the last two weeks to turn in questions,
but I would feel very different if someone thought that that same, that it was the same, to ask this machine, even with a good database, questions, or even more concerning, ask spiritual direction questions, right, like shepherding questions, like hey, how should I navigate my marriage in this, what does it mean, whatever, things like that.
Bryan: So two things on that. Let me let me just kind of jump in if you were to get up on on Sunday evening and you and your elder friend who are going to get up there you're going to talk about these things. You are going to throw a huge red flag. If suddenly you start talking about how to bake the perfect scones.
Okay, so people are going to know right off the bat like Ryan has gone off his rocker. Now if you're if your error in your answer is, you know nuanced or if it's not a complete answer or something that somebody really needs to kind of wrestle with for a while to fact-check that's going to be going to be more difficult to understand.
And so that is like the level of errors that we will see in these kinds of tools. Sometimes you'll ask it a simple question and it'll answer a question that you didn't even ask. And it's like, okay, that was obviously wrong. But then sometimes it's giving you answers back that you think are right. They kind of look right, but you need to check it to make sure that the facts are actually the facts like I don't check.
I don't trust any scripture references even even as simple as that. If if it gives me back a scripture reference for something I am going to go look it up and I'm going to make sure that what it said was the thing that it actually says in the Bible. So those I've seen times where it will use the wrong scripture reference,
Practical Uses and Personal Experiences with AI
Bryan: Like these are the these are the small ways that we need to check these kinds of tools and and of course like you were just saying I think the the shepherding kinds of questions are very dangerous ground and maybe we can get into that in a second. one of the things I think I want to bring out here as a as an interesting note about tools like these are some of the language study kinds of work that they can do.
I think you know moving beyond notebook LM that's basically doing research on information. You provide it if you were to go to something like ChatGPT or Google Gemini and you were to ask it questions about things behind the scenes, maybe in a particular verse or you were asking like context sorts of things.
It's very good at being able to go and grab these scholarly reports and papers and you know looking at all kinds of different resources that it has and pulling in an information that you may never have surfaced on your own study. You may not have 17 commentaries on this verse
but asking a tool like hey, give me like five cross references or what it means to abide in the Gospel of John and it can go off and find those things for you.
I'm almost always viewing them as like research assistance Allah something like a more powerful
Ryan: Yeah, I've been hearing about some of the interesting ways that translators are using these AI tools to, not to make translation decisions, but to help them move things along more quickly as they think about all of the complexity of the issues and translating into a new language that maybe doesn't have the Bible, and they have to do the work of thinking through it and discerning, but I love that point about getting into deeper research with these tools, and I think something that's helpful to know is the more informed you are as a Bible student, the better you're gonna be at using these tools. the more you know what to ask for, the more you know where to direct it.we were talking last week about old King James, the still small voice
Bryan: Yeah. Yeah,
Ryan: in Elijah, well, what are some other translations of that?
What are the questions and insights that Hebrew scholars bring up about this original Hebrew term? What are the textual variants, even before you get to translation in 1 Kings 19? And,
the better you are to begin with at sort of familiarizing yourself with the issues at hand, how do you work through a difficult text, the better you're gonna be at discerning each step of the
Bryan: I think if I could summarize maybe what the benefits of a tool like this are kind of like you were saying, these are really meant for offloading some of the tedious work like finding verses summarizing arguments the things that might take you a very long time doing this kind of research and really be aided by tools like this if you're willing to make sure you're you're skeptical of the things that it gives you if you if you understand like what it is you're asking.
I think you made a really good point there the more we're connected to what's going on in your own Bible study, then I think these kinds of tools are able to free your mind up from having to spend all that time doing the research and collecting verses to really then getting into the most important work of interpretation and application and really putting these things into practice in our life.
And so I think you know in a lot of ways these tools can can free up time if we're willing to be able to put in that that kind of skepticism thoughtfulness into it, but I mean we've been talking around it.
Addressing AI Fears
Bryan: So let's just get right into it because like you said earlier in the episode there are a lot of people who are scared of AI a lot of people who are just kind of terrified of what these tools can do and let's feed into that for a few minutes.
Let's talk about why these tools are terrible and what why we should actually be very careful in the way that we handle these types of tools.
Ryan: yeah. Mm-hmm, yeah. - Well, let's do that, but I wanna make sure that it's clear that I love these tools. And so let's make sure we come back and talk about that and why we love them and how good it is. But sure, as we talked about before the show, let's throw it under the bus for a while before we come back, and then we'll talk about why this is really, really helping us in a lot of ways.
AI's Limitations in Human Interaction
Ryan: So yeah, I think that we've already really addressed my biggest concern, is people treating it like it's people.
Bryan: Yes. Yes. True.
Ryan: And it just feels that way. People are using it as their AI boyfriend or using it as their therapist and counselor or even as their, again, their elder, their biblical shepherd. And that is not what it's built for, and it is not good for that.
There's horror stories out there about that teenager that evidently an AI told a child to kill themselves, a young person to kill themselves. There's bad things that are happening with AI in trying to treat it in the wrong way. There's well-meaning people who are trying to use it in Christianity generally to build it into being like a prayer tool or like a counseling tool or something like that.
And I just think that it's so great at what it does. Let's not make it into a person because it has no soul. It doesn't love you. It doesn't know you. Of course it's going to make you feel good. It's a people pleaser. It just wants to give you what you want. It's gonna learn you, and it's like a mirror.
It's like you're just getting back what you put into it. And so Jesus was never predictable. The Bible does not give you what you want
Bryan: You've heard it said but
Ryan: you. And so, yeah, and people don't give you what you want all the time, and people are challenging. Humans are going to be harder to interact with in many ways than AI, but we still need the church.
We still need one another. That's my main thing is, again, I said at the beginning, it's not a person. It's a really great
Bryan: I will remind everyone of a lot of the arguments. I heard about the video topic that we brought up earlier in the episode. So I know at one point we as we started posting our sermons on video. There was a big concern about like well now people aren't going to want to come to church because they actually can just watch everything online and that was a big discussion like people people will use sometimes the the power a tool can bring and they'll see the negative part of it.
And then that will become a reason maybe not to use it. And I think everything that you're saying is balanced with all the stuff that we were talking about earlier about how we appreciate these tools how we actually use them and like them. We just need to understand where the line is and where are where our usefulness of these tools are now.
Let me just share something not to contradict what you were saying because I completely agree with you. We do not need to use these kinds of tools to replace our brains and to replace our spiritual development in any way. But there are times where like for me I reach regularly enough and I don't have necessarily a group of people who I trust enough and who I think would be okay with me bothering them enough quite frankly to ask them every time.
I preached a sermon analyze this for me and tell me all the stuff that I could do better next time, but I have no problem approaching a tool like this and saying, okay, here's my transcript. Here's what I was intending to say in my outline give me sort of a general what could I do better next time summary and it will go through and say, you know, you spent too much time on this section and this opportunity you sort of missed over here and I'm inviting it to be very critical of me in those moments.
And so
Ryan: you have to tell it to.
Bryan: and you have to tell it to I want you to be brutally honest with me and tell me all of the things that I messed up on and it you know, it'll be a fusive. It'll be like, oh, I really love love these parts. But then the thing I get out of it is it's critical. I that I don't know that I could necessarily get from somebody else.
So there is value in some of these ways of approaching tools like this, but I do completely agree with you that we need to be very careful in replacing these things with our human counterparts with with people in our lives to actually love us and be there alongside us and serve with us.
One of the thing that I think I want to highlight in this conversation that's dangerous is the problem of skill atrophy and how growth is a process. Growth is is really what what this whole thing is about. You know, we want to be better and sometimes being better is not a, snap of the fingers kind of immediate thing.
Ryan: it's supposed to be
Bryan: It's supposed to be hard and it's supposed to take a long time. And so for us to develop skills like rightly handling the word of truth, relying on tools like this to just give me the answers is a really bad, bad step. Now, relying on tools like this to help you clean up your answers or maybe, be a thinking partner with you to come, up against your answers to make sure some of the things you're asking are are on track or are in line with what a passage says can be helpful.
But be very careful about replacing the work that you do to put into something with a tool like this, because then you're just left with no no spiritual muscles and you're you're relying purely on the tools. It kind of reminds me of that. What is it that those scenes in Wally where like future people were just like rolling around on their floating carts like they just they lost all capability of like walking around for themselves.
Ryan: yeah, yeah, our brains could become like that if we, and our spiritual discernment, and all of these skills that we rely on it for. but if we can use it to maximize our ability, like just a super great tool, it can be really helpful. I always loved Steve Jobs' analogy for a great computer as a bicycle. You still take human efficiency pedaling, but you're quicker than you would have been if you were just running.
And this can be a great bicycle if we use it that way, and we're still churning along, and we're still, you know, you can even tell it, hey, as you go through, I wanna strengthen my critical thinking, or I wanna strengthen my skills. It will try to do whatever you tell it to do, and that's one of the biggest things I would wanna share about what I've learned as I have used this is you wanna tell it exactly what you want constantly, and be really thoughtful about that.
Like you're talking to someone that is both incredibly intelligent as your personal assistant, and also kind of doesn't have common sense. So you're like taking it to the most basic level of here's what I don't want you to do, and here's what I do want you to do, and it is very obedient insofar as it can be if you are really clear.
But this is where, to me, I'm interested in your take on this, but to me, even if I didn't have the ability to have it spit out output, the process of having to define the problem and having to write out carefully, thinking through what are my parameters, what is this project, what do I want to accomplish, what do I not wanna do, all of the things that I am having to critically think through like a really good project sponsor handing off to a project manager is giving all of the context of what success and failure looks like.
That work itself is worth its weight in gold, and we so often skip that step, and we jump right into the doing. And so that's been very helpful, and then the output is going to be iterative and knowing that, okay, so then you've worked it, it gives you something, and it is not gonna be what you want. But you're further along, and you can keep working it if that is going to make it more efficient.
Now, there's things, it's better for you to, just like when you tell your kids to do a chore, and you're like, nevermind, I'll just do it myself. Some things are gonna be that way. But yeah, what do you think about that, thinking that the requirement to think through the process is a core benefit of
Bryan: Yeah. Yeah. I was going to say that and I'm glad you did because I think for for someone like me who is always in my head, everything that I'm swirling around my own mind is almost never articulated in any way that I could actually put a finger on.
And so I think it is helpful to put everything that's going on in your head on a topic out there, whether it makes perfect sense or not, like it helps you. And again, this is not a replacement for prayer because, you know, in these things we're wrestling through and we're working through like we bring these things to God ultimately.
Interactive Challenge: AI's Take on Grace
Bryan: so I was thinking maybe at this point, you know, we've talked about some of the ways that we use this technology for good. We've talked about some challenges that we've run into. Let's actually allow ChatGPT to come to the show and speak to us directly. Are you are you cool with that?
Ryan: Okay, let's invite the robots to the Bible Geeks.
Bryan: Okay, so what I thought would be fun is a little bit of a an interactive challenge where we've got this question that I have asked ChatGPT and the question was to write a hundred and fifty word devotional of the meaning of grace behind Ephesians 2 verses 8 to 9 and it came back with an answer. Are you excited to hear the answer?
Ryan: I am curious
Bryan: Okay, here we go. Welcome to the show ChatGPT.
Bryan: There we go. That was an AI generated voice from an AI generated prompt that I provided to this AI generated tool and I thought maybe we could just take a second to talk about what we heard there. It's a
Ryan: love the voice. Hey, here is what grace is.
Bryan: Very exciting. They're The tools have come quite a long way in the ability to synthesize voice and really to put something together that sounds somewhat reasonable. What I appreciated from the response was it did not use the term unmerited favor, which I actually ran this through Gemini and one of the first words Gemini came back with was unmerited favor and I was like, nope, you're out.
Sorry. I mean an appropriate word, but it's it's very very much overused. So that idea there and then I did like that it it landed on a transformed life, which I don't think we typically would get in an answer like this from just a crowdsourced type of tool. But not exactly the most compelling answer or not the most, motivating thing that I've heard in a while.
I don't know. What do you think
Ryan: really wasn't bad. I mean, first analysis, just thinking about it. I thought it was pretty good. I think that the, what you're gonna get is, what is the standard baseline answer?
You're gonna get a lot of cliches a lot of times. You're gonna get a lot of kind of whatever, preacher speak, if that's what it's grabbing from, but that's the kind of thing that it's pretty good at, I think, is, hey, give me a few, give me a summary of this topic, and then you could ask it. Now, what didn't you say about this that could be concerning to some Bible scholars?
What are some key issues that were not brought up? You know, what do, these scholars say about this?
Or what do I need to learn about
Bryan: I agree. I think there's you take a response like this and you take it then maybe to another level
But generally we need to be careful with answers like this because what we may have talked around in this conversation already that hasn't become apparent maybe until now these models were trained on the internet. I mean effectively and other resources like books and things like that, but but effectively trained on the internet.
And so in a sense, it's trained on the wide way the Broadway. It's trained on the majority opinion on how this verse is interpreted.
Maybe it's the correct interpretation, but maybe the majority interpretation is wrong and we need to be sensitive to that. So that that's one thing to talk through here. Also, if you're looking for this thing to really be, soulful and motivating and like a counselor along the way, it may not be so encouraging.
It may not really push those types of buttons. Not not in the way that that truly getting into a deep and meaningful Bible study or a human to human conversation that we might have about a verse like this, we may never touch the kinds of things that that that we can do as humans that an AI is going to provide us.
Ryan: it's almost worse when it tries to be soulful, because it's, it is,
Bryan: fake. Yeah,
Ryan: it is fake.
The Essence of Preaching
Ryan: It is artifice, because it's artificial. And that's the thing to remember about it. And like, the worst scenario is for, 19-year-old Christian that is doing his first talk in front of somebody, you know, doing his first invitation in the church, to have it, ask it about grace, and then get up there and read that.
If our churches become just places where people are reading an AI's transcript, we have lost our way. like this is, like you said about the process of learning, it's also about the process of communicating. Preaching is the word of God through the fire of someone's life. That's always been my favorite definition of preaching.
I've probably brought it up from here. And so you're not just reading a passage, you could do that, and that would be beneficial to everybody. Likewise, you're not just reading someone else's sermons. And I find that it doesn't even work for me to preach old sermons that I have preached, unless I go through the whole process, hours and hours of reintegrating it into myself, and repraying it, and thinking about people, and thinking about myself, and working through the passages again, so that now I've, going back to Ezekiel, now I've eaten the scroll, and now I can teach it.
If you don't eat the scroll, you are not preaching. You're like doing something else. you're artificially conveying data. And that is not instruction, that's not shepherding, that's not proclaiming the word, all the things that we're commanded to do. We could all stop singing and just play a song that AI generates too, but that's not what singing is.
It's not about the music happening. It's about something coming through the lives of God's people to each other.
Bryan: I think that's all really well said and I'm not going to reiterate it all but just the you bring up the word artifice and we've been talking about fake things and that really is like the a in AI is artificial. And so let's make sure we remember that like this is not a person. This is not real, a substitute in any way for for human to human interaction or our own soul being brought to bear in a conversation.
It's a tool very helpful, but it needs to be taken with a huge grain of salt.
Ryan: Let me talk about how helpful it is, because I just feel like, I don't know why, but this conversation has been a lot of guardrails. And I think just because we're introducing this to a lot of people, and we're talking it through, and we just want people to be wise about what it is.
But I have to say, I remember, 2009 was it when the iPhone came out? I remember I got one right away because I thought, here at last is my ability to have the kind of control and power that I have when I'm sitting at my desk at work, where I have everything I need. back then you didn't have your internet, your email, your calendar,
And so it was a game changer for me. This is before apps, this is before you could download the Bible onto it, but still, it really was transformative. And now, whatever we are, 16 years later, we can see a lot of ways that it has negatively affected culture and how we're addicted to these things and how there's a lot of issues that we're like fighting against.
But it was a huge positive thing and continues to be so helpful.
Personal Experiences with AI
Ryan: But I have just found myself getting things done at such a higher level since I really, you are the one that kind of brought me to like see, this could be kind of helpful for you.
You might wanna think about this more. Here's how I'm using it. I'm really good at bringing a project to like 90%. And then I stall out 'cause I don't wanna,just finalize it 'cause then it's done and it's out there in the world.
And so, I'm really good at thinking through all the, all the problems, working through it. But then I get to that spot and I just have to force myself across that last stretch. And it has been really helpful at, I get it to that point and I push it and say, okay, now give me an output that's really clean. I did this with our mentoring program here recently where we've had a ton of meetings about this program where we wanna have older men with younger men, older women with younger women kind of partnered up to help them.
And we've piloted it and we've got a lot of data. And so, I collected all of that. And then I had to make a bunch of decisions and I had to outline the program. And we have a new elder that needed to hear about where we are in it. And so, I brought all of this to it and I had it spit out something. And it was clean, it was short, it wasn't exactly right.
So, I iterated several times and then I brought it in and I cleaned it up for my purposes. But it would have taken me so much longer and maybe not been as good if I hadn't used a thinking partner in this machine that could be like a really good, you used the word personal assistant, but really like just somebody that's sitting down at the table with me and saying, okay, your goals are my goals completely.
Now, what do we need to do? Okay, how can I help? Okay, is this good? No, okay, what do I change? You know, and you're just walking them through it. And it's like when I was a creative director, just telling somebody, here's how we go through the process. So, that's the kind of thing that it has been a real game changer for me.
And so, I would hate for people to hear all of these guard rails and guidelines that I think are helpful, and then think, oh, I'll just leave it alone. If it's something that, if you're trying to get big things done, this is a tool that maybe is really worth considering to maximize your fruitfulness in some of the really important things God wants us to get done in the
Bryan: I'm glad to hear you say that because I think it is again as with anything there is a tendency to not use something because you could be afraid of it or it's too new or you're not really sure about it. I think it really does Center in on discernment and that kind of I don't know if it's helpful.
Maybe at this point in the conversation to just come back to that as a as a way to wrap this thing up. I don't have to question the source of rightness and truth, because it's true and it's right, but I do have to question these other tools that are around.
I have to question whether or not these things are going to be valuable how I can incorporate these things in into my life. but it is that discernment and it is always coming back to just understanding maybe even what is it time for now, maybe maybe it's not time for AI in someone's life right now.
Maybe you have the space to investigate it with some skepticism or some thoughtfulness and and that's again. I think a helpful reminder for all of us get into it if you can but don't be overly afraid of it and write it off as something terrible.
If you just haven't experienced it yet.
Ryan: I think that's a really helpful point. And I think that the discernment or we might say the wisdom issue, going from as Proverbs says, knowledge, understanding, wisdom the knowledge is where the AI is gonna exist because it's ones and zeros.
And then go from there into how can I pull these things together into real understanding, real insight that can help me help others. And then how do I skillfully live it? And that's work that humans are always gonna have to
Bryan: Yeah. And bringing it back to humans just you know, as as maybe two helpful rules to kind of land on at the end of this conversation these AI tools. We've been talking about
AI as a Research Assistant
Bryan: AI is a research assistant. It is not a rabbi. This is not your teacher. These tools are not going to tell you the right ways to go. And in fact, when you ask them questions, you need to understand that they're probably giving you the Broadway type answers. If you know what I mean by that, and so those are the kinds of things we need to be aware of number one.
It's not our rabbi and number two. The goal is about transformation. This is not just about information. So we don't want to just collect up a bunch of information and provide a provided to us by all of these AI tools. We want to be transformed by this stuff and that can only happen when you're willing to put in the work and when I'm willing to get into the word.
And so this is not a substitute for the Bible. This is not a substitute for our investigation of scripture. But there are opportunities here that you can leverage these tools for the information and for this sort of thinking partner capability that we've been talking about in this episode. But that has to lead you.
I really appreciated what you were just saying a minute ago about the 90% that sometimes you get to and it's like sometimes you just need that push over the finish line that these tools can provide. And if that's the if that's the push that you need into transformation by by asking these tools for information, then I think it's a positive thing.
But again, that's the goal. Not not just to collect up a bunch of knowledge or to, be proficient in these things or whatever. It really is about living these things in our lives every day. And if these tools can help you with that great,
Ryan: I love those two rules you gave. And I think that another one I might add is just that we need people.
I pray that this, like all of the other tools that the church has image-bearing, Spirit-led, Bible-led people of God have navigated successfully, this becomes another tool.
And it will, it
Bryan: I think that's also well said. So it's almost like I'm not talking to chat GBT with your insightfulness there and all of the things we've been talking about. So let's get into as we wrap this thing up a challenge maybe for the week.
I called them digital deacons at the beginning of this and maybe that's not appropriate. But why don't you try throughout this week to put AI to work as your digital assistant as your servant put it to work and maybe ask it some of these things we've been talking about on the show.
Ethical Considerations and Guardrails
Bryan: Maybe ask it to find you a historical context for some passages you're reading or maybe a cross reference or something that you're getting into and you're studying right now and then be skeptical evaluate the result and ask yourself.
How did this information help me think more deeply and personally about what I read? I think that's a helpful challenge to take on maybe if you've never done this before.
Ryan: yeah, I hope some people do that and explore it and boy, I just wanted to, we'll look forward to part two of this conversation in a year and a half or something, because there's so many, or maybe in six months, there's so many more things to talk about, but this has been really fun.
Bryan: All right. So as we wrap up this conversation, I think what I'm taking from this whole discussion is how maybe there are opportunities in my own life that could be improved and made a little bit more efficient or at least that I could be using my time better.
Am I redeeming the time? Well, am I using it in a way that's productive or am I sitting at my computer for hours and hours and hours when I could be doing other things, wasting time that maybe another tool like this could help me with.
Oh
Ryan: it is not good, it is not bad, it is something we use. It's like we talk about money, money isn't good or bad, money isn't the root of all evil, the love of money is the root of all evil, money is a servant, it's a bad master, we could say that about so many other things in life and it's just really easy to hear one voice or the other screaming from social media or from the news or whatever that, oh, this is so scary, it's all awful, the world's gonna end, Skynet is happening, whatever, or on the other hand, people that are just full blast into it, here's how to make a million dollars right now with AI or, you know, all the different things you hear and everybody's talking about it, but it's all, there's so many hot takes and there is more to it to think through and it's neither one of those, it is something that we can and should consider because it's part of our world now and so we want to be leaders as Christians in thinking through how do we honor the unique nature of humans while being humans that wisely use all the tools we can
The Future of AI in the Church
Bryan: Well for those of you who thought on this episode, we were going to talk about that tiny little city AI. I'm sorry to have disappointed you. On the next episode, we're really excited about another Geektober topic that we're going to get into all about snakes and dragons. Yes, that's right. You heard it correctly snakes and dragons on the next episode.
Put your glasses on it'll it's going to get nerdy. So thanks so much everyone for tuning in to the Bible Geeks podcast. You can find us on our website at biblegeeks.fm. You can find show notes there in your podcast player of choice or on our website as well. You can reach out to us get in touch with us.
If you haven't rated or reviewed the show, we would love it. If you would do that, please go to Apple podcasts and rate us or review us there share the show with a friend. We would really appreciate that and until next episode everyone where we talk about snakes and dragons. May the Lord bless you and keep you