Well. Welcome, David Thorne. How are you doing, buddy?
Speaker3: [00:00:14] Doing good.
Speaker2: [00:00:15] All right. I’m pumped about our our talk today. Um, because for our audience, having David Thorne on the show is just a delight. David’s on our team here at Missional Marketing, and he gets to do a lot of cool things. I actually wish I was him because he gets to do cool stuff and he lives in that, like, like R&D world here at missional marketing. And I love that stuff. So I’m really pumped about what he gets to do. And I’m excited for this talk today because, David, you went somewhere recently. Let our audience know a little bit. First of all, just your role and then where you went recently that we’re going to talk about.
Speaker3: [00:00:58] Yeah, yeah. So mission marketing, I get to play around with all the fun ideas, believing that what if all things were possible and then figure out, how do we do those things to help our churches? And recently, another coworker and I, Jason Collier, we went to San Francisco for the inaugural conference, which was the first one that that they’ve had where they invited, you know, Nvidia, Google, anthropic, all the big players hugging face, anyone you could think of. They were there. And we just got to listen to them talk about AI, which was awesome for two days.
Speaker2: [00:01:34] I’m sure people are going. That sounds like a total bore. If you’re not into that. But for others, they’re going really? You got to go to oh my goodness. So tell me just a just we’re going to dive into some questions that we wanted to ask you about your experience. But what are a couple of highlights from from being there just at the conference.
Speaker3: [00:01:56] Yeah. It was cool to actually like went up and met the the co-founder of anthropic which Claude, which is their large language model is like my favorite large language model. So it’s really cool to kind of geek out for just a minute and say your large language model, Claude is so amazing. And we got to see the the bigwig from Google and you know, yeah, like Nvidia and these other ones and go up and just talk to them and be like, hey, tell us what’s going on in Nvidia. And that was it was kind of cool.
Speaker1: [00:02:32] I’m going to press pause already okay. Because we had this conversation before we got started. So we don’t want to get too technical in this podcast episode because our primary audience is comms directors and comms people. You’re using technical language already, David. Large language model. Break that down for us. What is a large language model when it comes to artificial intelligence? Yes it is.
Speaker3: [00:02:54] Basically, if you think of it as like a brain and you kind of put a bunch, you feed it a bunch of cool stuff and then that brain can think things. So it’s, it’s like a, um, a brain of words and thoughts and ideas and facts and research and information. And then you get to use that brain and ask it questions.
Speaker1: [00:03:18] Okay. And the most common large language model that people have probably used is.
Speaker3: [00:03:23] Claude GPT. Chat. Chat.
Speaker1: [00:03:27] Chat GPT. Yeah. Claude. Your favorite. See? That’s what that’s what parent. That’s what parents accidentally do with their favorite kids. They call them by the wrong name. Your favorite is Claude. My favorite is Claude. That’s the one I’ve used most. But ChatGPT would be the large language model that most people are probably familiar with. There’s a few others, like is Bard. Is that also a large language model? Is that the same, same idea?
Speaker3: [00:03:49] Bard is Google’s.
Speaker1: [00:03:51] Okay. All right. So there’s a few of those out there. So yeah, I’m just you know what? I’m just trying to I’m not super, super deep on the technology of all this stuff, although I’m using some of these tools. So I just want to make sure that we, we, we let everybody keep up with this so they don’t turn off the podcast episode in the first two minutes. Okay. Let’s continue going. Sorry. So you got to geek out and you were pretty good at that. Guess.
Speaker3: [00:04:16] It was fun, I liked it.
Speaker1: [00:04:18] He didn’t answer the question, though. I was hoping he would. You got to geek out at a conference. And you were pretty good at geeking out. Yeah.
Speaker3: [00:04:25] No, because I’m not a geek, so.
Speaker1: [00:04:27] Okay. All right, all right, all right.
Speaker2: [00:04:30] But you’re really creative in that, in that environment. And there’s a reason why David’s kind of heading up our R&D team is just because he thinks that way. He loves the local church. He’s a follower of Jesus. And he’s like, how can I take technology in the church and bring it together? So that’s what’s really fun. Okay, David, I have a couple of questions. Now. Obviously, this is this is beyond the conference. You just got to go to the conference is amazing. But we’ll dive into that a little bit. But. Okay generative AI.
Speaker4: [00:05:05] What is.
Speaker2: [00:05:06] Generative AI? And you said it’s kind of like a calculator for language. So what is what is generative AI?
Speaker3: [00:05:14] Yeah that’s a great, great question. So one of the interviews that was like one of my favorite quotes of the entire conference, and where a guy said that when it comes to AI stuff, that AI is no more than a calculator for language. Now he was talking about generative AI, which is what you’re asking. Generative AI is basically what all of those things like ChatGPT do is they basically come up with unique thought and answer questions as if you asked a friend. So if I ask you, Jason, Jason, tell me about the weather today. You’re going to generate a response. Yep. And that response is probably going to sound very human like. And that’s exactly what things like ChatGPT clawed, barred, that’s what they do, is they come up with human like responses. Yes, Mister Bert.
Speaker1: [00:06:10] Raised my hand. For those of you who are listening online, I raised my hand, which is why David called on me. Okay, I’m a student in class here, so would know a lot of people are using tools that create images. Is that also gender generated? That’s a hard word to say. Jennifer. Generative AI is that also generative generative AI where you’re like putting in a prompt and it’s providing you like graphics or images and things like that.
Speaker3: [00:06:35] Yeah, it’s basically it’s it can be anything. So you’re putting in a, a question asking a brain to do something for you. That thing it can do can be can it can generate words, written words, it can generate set out loud words. It could generate an image. So anything that can be generated could be considered a generative AI.
Speaker2: [00:07:05] You know. So that probably is catching people off guard. That might be listening to this because you could think, well. Then this has been the scare. This has been the unknown question about is this going to replace humans in terms of what we do? Is my job in jeopardy of being replaced by AI?
Speaker3: [00:07:26] Yeah, that is the question. And here’s one thing, is that one of the guys that spoke, he he worked in the Obama administration for I stuff, and he was just kind of saying that unequivocally, that’s the wrong question to ask, because any I should never replace humans. It should augment what we do. And so when we think about what tasks we want to do, the question is never, how can I get this? How can I get the AI to do this instead of me? It’s saying, how can I get the AI to do what I need to do better and faster than if I did it myself? So that’s where if you think of AI as like your language calculator friend, for all the crazy tasks you have to get done in a week, it becomes a really helpful tool. Instead of a scary monster, it’s going to take your job.
Speaker2: [00:08:28] Now, I know this to be, um, kind of a maybe a dumb question, but when I think about right, how are you using AI right now in the real world? I look at my iPhone. Okay, so whenever I want to turn on my iPhone, I just look at it. It reads my face and it turns on instead of me having to enter a password. And that’s a kind of a dumb example, but that’s AI helping me just be more efficient.
Speaker3: [00:09:00] Yeah, yeah, that’s a helpful thing. Yeah.
Speaker1: [00:09:05] Raising my hand. I’m raising my hand again. Okay, so. So I’m again. I’m just trying to. Trying to drill down on what you’re explaining there. I should augment humans, not replace them. So you’re basically saying it should actually help us do our jobs better, more efficiently, quicker. Here’s the first example that comes to my mind. Tell me if you think this is on the right track. One of the new features in Photoshop is generative fill, right? So you can take a picture and you can if it’s not completely there, you can use this generative fill tool to fill in the picture or to change something about the picture. And in the old days of Photoshop, meaning like, I don’t know, six weeks ago an artist in Photoshop could actually do that. They would just have to do the whole thing manually, draw it all out by hand. It would have been a very painstaking process. And there are some Photoshop people who are like, amazingly, brilliantly gifted who can do that kind of stuff. But now the AI in Photoshop actually does it for you, right? So what might used to take you hours to do, you can now do in a matter of a few seconds. So is that I think that is that a good practical example of augmentation versus replacing.
Speaker3: [00:10:17] Yeah. Yeah. No. I’ll give you another perfect example is transcription. So you know, let’s say uh, back in the day you had to physically write a transcription or even like write it, have have some kind of machine read it out, and then you’re going through and fixing everything. The generative AIS now are getting better and better and better at figuring out how to do transcription. Great the first time. And then you could possibly use a second AI to say, hey, take that transcript and fix all the errors in it and you’re done. 15 minutes done. And just saying, like there’s all kinds of those examples.
Speaker4: [00:11:04] Interesting.
Speaker1: [00:11:04] Now, I would say in that case though, it could actually replace someone’s job. If someone’s job is to transcribe something and edit. Right. Because now you’re using tools that do on a technical standpoint, kind of replace what the human being once did. So that means that that human being who used to be the one that transcribed the content or did the editing of content, there’s a new skill that they’re going to need to learn in order to, you know, stay relevant and stay employable, I guess on some level, sure.
Speaker4: [00:11:35] Yeah.
Speaker3: [00:11:35] But what I might add on to that is, say, for that person who does that, I bet you that the church or whoever does that has a list of 20 tasks that have to be done every day and every week, and the last five never get done. So what would happen if you knocked off? You knocked off the first three and it took them a quarter of the time. What will happen is, is over time, those last four things will now get done. Although then the church person is going to add five more. So that’s that’s how these things work.
Speaker4: [00:12:09] Yeah.
Speaker2: [00:12:11] So, um, it seems like, you know, all we hear about these days is, is I, you know, you’re seeing it on commercials all the time. You mentioned that, right? It’s just everywhere. And yet it’s really not in our lives. I mean, some of this is in our lives, but I think you made a statement here that it’s it’s kind of more technical now, but it’s going to become more accessible over time. What do you mean by by that?
Speaker3: [00:12:36] Yeah. So we’re at this weird moment where their stuff is being created so fast you cannot keep up. Um, like, for example, just in the last week and a half ChatGPT for vision can like see what is in an image beyond just creating an image which didn’t exist two weeks ago, just for example, what’s going to come out next week? I don’t know, but, um, uh, it’s it’s just saying that, um, there’s a lot of things you can do that’s very, very deep, but it does take some a smart developer person to do it. Um, with that being said, there are new products and companies that are being blowing up every single day. Like, say you need task done. Guess what? There’s a company that now uses an AI that does that. You need task be done. Guess what? There’s a company that exists right this second that can that can do task B for you. That’s an AI company. One of the challenges that church leaders are really going to have to wrestle through is where do I spend my money? I have a little bit of money, I have a lot of tasks that I need done, and I would like AI to help me do those tasks faster. Um, and so there’s a there’s a solution to that, which is basically saying that every church leader, pastor, communications leader should start by writing out a list of what are all the tasks that you do in a week. Take a piece of paper or or your computer and write out all the tasks that you do in a week and then say of those tasks, what takes the most time or is the most challenging for you to do? And for those tasks that take the most time or the most challenging to do, those could be places that you could consider finding an AI that could make those tasks less painful and take less time.
Speaker2: [00:14:50] Interesting. So when you’re at this conference and you’re obviously thinking about the church, you mean this is what we do every day? All three of us get to work and serve churches. What’s it going to? What do you think is going to what’s it going to be like for the church moving forward? How is going to help us do ministry and reach more people for Jesus?
Speaker3: [00:15:13] Yeah.
Speaker4: [00:15:15] Well.
Speaker3: [00:15:16] One one caveat I just want to mention for on that question is I think it’s really helpful for our church people and leaders to be aware that AI is going to become more in the background more and more every day, where right now we’re like, hey, I. But pretty soon AI is going to be built into every single thing we do to such an extent that we won’t even know it’s there. Um, so every single platform we use, every internet browser, every single program we use, it’s all going to include AI. So with that, when I think about what the church it’s saying, this is not a technology that we even have the ability to not use. It’s a matter of are we going to choose to to to lean into it and use it to help us do what we do better, faster than we’re doing it right now? Or are we going to wait until it’s already built into every single thing we do? It’s not a matter of if, it’s just a matter of when, and we can choose to do something faster and better now. Or we can wait and it will be built into the things we do eventually anyways, thrust upon us.
Speaker2: [00:16:31] Yeah, yeah. I you know, I talked a lot of pastors. I asked one pastor tell me about how how do you it’s a younger guy. So you know, he’s a little bit more open, a little more technologically advanced. I said, tell me about where are you on your stance of AI? And he said that, you know, there’s probably about a third of the pastors are all for it, a third that are unsure about it, and then a third that are just scared because they don’t know. And so my question to you, David, is, I guess when it comes to the ethics of AI, what would we say? We’re in the Wild West right now. It’s like anything kind of goes, explain what you explain to our audience about where are we in this ethical ness of usage of AI?
Speaker4: [00:17:19] Yes.
Speaker3: [00:17:20] Are you laughing? Because that may or may not be be a word.
Speaker1: [00:17:23] I am laughing because ethical ness may or may not be a word, but, but, but but you know what? But had a college professor once that said, if it gets used and people understand what it means, it’s a word. So I think we understood what Jason meant. Ethical ness.
Speaker4: [00:17:37] Yeah.
Speaker3: [00:17:39] Um, so. Here’s some good news. I’ll say this. All of the big boys that are out there like anthropic, Microsoft, Google, IBM. The good news is, is they are wrestling with this now partially because they don’t want to get sued. And so therefore their motivation is I don’t want to have to give out tons and tons of money due to our lack of ethics. However, that is good news because they are being forced to to talk through of how do we make our AI that we are using harmless, helpful and honest, which is really good. Um, the other thing that we need to understand is that right this second, there’s no clear cut laws or rules on how, um, AI trains its information. So kind of like how do you train the brain? You know, the brain knows stuff. How do you train your brain? And that’s what one of the biggest challenges is, let’s say, you know, we have an image generator. And this is one of the lawsuits that are around right now. And we say, well, we need images so it knows what things look like. And so some places. Getty images is one of the the largest online repositories of images. And some image generator generation AIS were trained on like that included Getty Images So Getty Images is suing for their stuff now.
Speaker3: [00:19:20] Um, it’s what’s important to know, again, is that some companies are really taking this seriously. So like Adobe, for example, they have an AI, an image generation called Firefly, and they’re trying to be more transparent about, um, where they’re getting their images from to think through how to pay people whose images they are using, or whose images were used in the generation of an image and stuff like that. Um, not in America, but in Europe. The European Union is working on an act that is trying to come up with a standardization for things like transparency and, you know, where information came from and the ethical use of AI so that it is helpful and harmless and that kind of stuff. So it’s being aware, you know, fellow Christian person out there, that it is a bit of the Wild West and that there’s no laws. Officially, however, it is worth knowing that if you stick to the big boys and what they’re doing, the big companies out there, they are highly motivated to not get sued and therefore are working to be as ethical as possible.
Speaker1: [00:20:43] I recently listened to an interview with someone, or was reading an article about the use of AI content in books and written materials, and this is just an example that there have been a lot of red flags raised with people self-publishing books on Amazon that were basically AI generated. It’s AI generated content, and one of the first steps that Amazon has started to implement is that when you upload a book into the self-publishing, what is it? Kindle? It’s publishing on demand, whatever. Whatever the platform is called, you actually have to disclose whether or not your content is AI generated or AI assisted, right. There’s kind of a differentiation between AI generated content and AI assisted content, and that includes the content in the book itself. It includes the any images or the artwork on the cover. You have to disclose that now in the current framework, as an end buyer, there’s no way for you to see whether or not the content was AI generated or AI assisted. They haven’t gone that far, but they have started to require that those that are claiming copyright authorship of the content or creative authorship, creative ownership of the content that they’re disclosing. If there was some sort of of of AI assistant or AI generated content in that. So those are steps in that direction. And I just thought I would mention that because I think it is it is again, Amazon’s obviously one of the big players as it relates to most people don’t think about, you know, Amazon as being a content marketplace, but it really is. It’s you know, most of us we’re buying like groceries and spark plugs and windshield wiper blades on Amazon. But it first and foremost is a is a place where people buy authored content. And I know many, many pastors publish content. You know, writers publish content. So it’s one of those spaces that is is experiencing even some of those growing pains of AI there.
Speaker3: [00:22:57] Yeah, that’s definitely true.
Speaker4: [00:23:00] Yeah.
Speaker2: [00:23:01] Yeah, I think a lot of a lot of church staff people are still. They’re probably starting to to mess with it and play with it. What would you what are some give me? I know I’m putting you on the spot here. What’s one low hanging fruit I task I could do that would pretty much be in every single church.
Speaker4: [00:23:25] Yeah.
Speaker3: [00:23:26] Uh, how about use it to help you write announcements every Sunday? So you have three pieces of information about an upcoming event. When, where and who. And then you can tell it. I would use a Claude for this task. Um, but. And say, write me an announcement for our church service. Cool. There you go. There’s one. Uh, one. The second thing might be, if you have a transcript of your pastor’s sermon, you probably have to add some kind of summary, a two sentence summary, and also like a multi paragraph summary, a few different places. Somewhere in your website, you’re going to need a two sentence summary. We’ll upload the transcript and ask for a two sentence summary, a two paragraph summary, and then a general summary. That’s a second thing you could do. Um, if you have to write a church letter about an email about an upcoming thing, um, you could use it to two ways. One is ask it to help you to write that letter. A second is you write it, you upload it to a generative system and you say, please fix this for many errors and help me to improve my writing. And those are two ways that you can use it to help you to potentially write something, but at least ask you. Help it to help you to fix errors and make it a little bit stronger than what you maybe originally wrote.
Speaker4: [00:25:00] Yeah, you can improve.
Speaker1: [00:25:01] Improve readability and impact. Yeah. For sure.
Speaker2: [00:25:04] Yeah. Really? Claude, is Claude dot I one thing I’ve learned to this is kind of cool is I don’t think most people know this, but go to YouTube where you’re you load it up that sermon, and if you click on those three little dots that are right below the video to the right, it’ll say show transcript. And you’re like, yeah, you can click on that and then you’ll have to click a second time to actually see the transcript. And YouTube will give you the whole transcript of that sermon. There you go. No need to spend any more money on these tools that you’re getting. Your transcript. Do it right there. And now you’ve got it. That’s I’ve always encouraged churches to do that. I love the I love taking that. Transcript. Bring it over to Claude. That’s a really good idea to say crawl this transcript and give me two paragraphs or give me a blog. Give me a 1500 word blog post on this.
Speaker4: [00:26:01] Right? Yeah, yeah. Now, once.
Speaker1: [00:26:02] You’ve uploaded, I mean, you can, you can get a you can get a blog post, you can get a small group discussion questions, you know, give me six discussion questions or ten discussion questions from this. Right. There’s so much that you can do, which again is those are probably in a lot of cases, as you mentioned earlier, David, those some of those tasks are the ones that are like seven, eight, nine on the list. They’re the ones that we really want to get to. But sometimes they don’t become the priority and they actually become really simple to do. When you’ve got an AI tool that you can use to deliver.
Speaker4: [00:26:33] Yeah, yeah.
Speaker2: [00:26:35] Now, we would be remiss by also saying we have a, we created a our own tool. You want to talk a little bit about what we are offering churches and specifically pastors.
Speaker3: [00:26:44] Yeah. So we’ve developed a tool that’s called Sermon Spark which you can download. It’s able to you’re able to use it for free at Sermon Spark II. And we like to say every sermon needs to spark an inspiration and help with coming up with things like creative sermon titles, coming up with small group questions. Um, I have a we’re going to be launching a new tool on it soon, which I, we call it, and I’ll unveil this here because it’s going to be coming out soon.
Speaker4: [00:27:17] Drumroll. Yes.
Speaker3: [00:27:20] Uh, a topic generator, meaning what we’re doing with there is imagine if you’re going to preach on a topic and you’re just kind of curious, like, what are things I should be thinking about if I were going to preach on that topic? So it gives you questions to ask, things to consider. If you were preaching to a non-Christian on this topic, what are things you should keep in mind? So that’s an example. That tool that doesn’t exist yet that is coming out soon on Sermon Spark. But we have a bunch of other tools that are that can help you with research of like, what are Bible verses on that topic? Um, or throw out an outline generation generator tool. There’s all kinds of tools, and we have a list of like ten more that we’re working on that we’re going to be adding to Sermon Spark.
Speaker4: [00:28:11] Yeah.
Speaker2: [00:28:12] Wow. I’m I’m super pumped about that. And you know, what’s really cool is, uh, we’re keeping an eye on what pastors need or what churches need. And we just got our other eye over here on like, the technology advances of things. And we’re just going, how can we bring those two together? And again, we’re not trying to replace people. We’re trying to enhance what you do, maybe quicker. Imagine a pastor saving five, eight hours of his or her day and now they can do more ministry, right? Or just, you know, have more free time to like, mentally get sharp again. You know, I mean, there’s a lot of fatigue out there in the church world. So maybe, maybe AI is just going to help you have less fatigue. And that’s good mental health.
Speaker3: [00:28:58] And I think it’s important when we talk about some of sermon spark, that’s where pastors and churches are like whoa whoa whoa whoa. Are you replacing and the the place of the Holy Spirit? The question is absolutely not. It’s kind of saying you already use Google and you don’t go to your church every Sunday and say, now this this week’s sermon, I want to let you know that I, I found this story on Google. It was, you know, I was googling and I found, in fact, the Bible verses that I’m telling you today, I found those on Biblegateway.com. You don’t you don’t do that. You know, a pastor uses tons of resources to help their with their sermon writing process. The Holy Spirit then helps you take those resources you identified and pull them together into something that is contextually valid for your congregation about what is God saying to your people? So sermon spark, it’s important to understand it is a tool that is just a super charged way to get all of the pieces you need, so that you and the Holy Spirit together can work together with those pieces, and then weave together the sermon that God’s trying to tell your people.
Speaker2: [00:30:13] Awesome.
Speaker1: [00:30:15] Yeah. It goes. That’s the difference between, again, AI assisted versus AI generated. We’re never proponents of people saying I created this if they didn’t actually create it, if it was created with artificial intelligence. But what we’re saying is we want to augment what God is already doing in and through your life, in and through your ministry, in and through your preaching and your teaching, by utilizing the tools that you have, you know, access to that a matter of weeks ago weren’t available to us.
Speaker4: [00:30:48] Yeah.
Speaker2: [00:30:50] Yeah. That’s crazy. Well, David, thank you so much for taking some time out to talk to us about AI. And certainly this is a this is a topic we will keep talking about a lot. In fact, we have a podcast on AI. You want to talk a little bit about that.
Speaker3: [00:31:08] Yeah. So we are continuing to grow a podcast on the topic of AI for churches. We actually have a Facebook group for churches where we are talking about this stuff every single day, and then we are continuing and developing a podcast on the same topic. And I’m going to be hosting that podcast. So I want to encourage you all to come listen to AI for churches where we are going to dig deep. There’s about 20 more things I would love to tell you that we did not get to in this episode. We’ll talk about those things on that podcast.
Speaker4: [00:31:41] Awesome.
Speaker2: [00:31:42] Well thank.
Speaker4: [00:31:42] You David.
Speaker1: [00:31:43] I’ve been I’ve been typing up a list of things that need to go in the show notes for this podcast episode. Links to sermon, spark, and Claude. A lot of the conversation points that we discussed today came from a blog post that David actually wrote that’s on the missional marketing website after he attended the conference. And so make sure you check out the show notes of this episode, because I’ve got a whole boatload of links that are in there. And of course, the AI for churches podcast link will be in there. Cassandra Robinson got that started for us about five months ago or so, and we’re transitioning over where David is going to pick up as kind of the lead voice in that in that space, because he’s leading our company in that space. And we also have a website that’s on a subdomain of the missional marketing website. It’s AI, missional marketing.com. And there’s tons of stuff, resources and things that you can access on the AI missional Marketing.com platform as well. So David, thanks again for joining us. For those of you who are watching on our YouTube channel, great looking people here. A couple of bald guys, couple with glasses and you can pick who’s who, those of you who are listening online. If you haven’t ever subscribed or you’re listening to the podcast, make sure that you do that and have it. If you haven’t ever left us a rating or review, we’d love to get your rating and your review, especially if it’s a five star review. That’s really what we want. If it’s less than five stars, why did you listen this far? If if you got this far, you don’t want to leave a five star review. I’m surprised that you got this far. But you know how the algorithms of the interwebs work. If you leave reviews and leave comments, it helps more people find the show. So thanks again, David. Thanks for those of you who are listening in, and we’ll see you next time.