Technology is advancing at a breakneck pace. New tools are emerging faster than most of us can learn to use them, and it can feel overwhelming—even for those who love tech. In the last year alone, we’ve seen incredible leaps in artificial intelligence (AI), especially with the rise of large language models (LLMs) like ChatGPT. These tools offer powerful new capabilities, but as with any new technology, they come with challenges and warnings to consider.  

In the conversation below, Rob Davidson of Concordia Technology Solutions sits down with Rev. Bill Johnson to discuss how AI might impact church ministry. Rev. Johnson, a tech-savvy pastor, is optimistic about the good AI can do in the coming years. However, he urges churches to approach AI with wisdom and caution. Together, Rob and Rev. Johnson explore three key areas of concern—reliability, privacy/security, and anthropomorphism—and how church leaders can navigate the opportunities and pitfalls of AI in ministry. 

Reliability 

DAVIDSON: Rev. Johnson, let’s start with reliability. These AI tools can do some amazing things, but I recall you mentioning that it doesn’t take long for an AI like ChatGPT to be “just plain wrong.” Why do these AI models often get things wrong, and what does that mean for church staff who might want to use them?  

JOHNSON: That’s a great question. One thing to understand is that large language models don’t actually “know” facts in the way we do. At their core, they’re statistical models predicting probable word patterns. So sometimes an AI will produce an answer that sounds confident and very plausible, but it can be completely incorrect—what we call a “hallucination.” And it happens regularly. You can try to minimize it by crafting your prompt carefully (for example, explicitly telling the AI to stick to known facts and not make assumptions), but even then, you have to double-check everything it outputs. In short, if a pastor or church staff member uses an AI-generated answer for something important, they absolutely need to fact-check it against reliable sources before trusting it.  

DAVIDSON: Given those limitations, have you found any ways that tools like ChatGPT can still be helpful in a church or ministry context? In other words, where do these AI tools excel despite not always being reliable with facts?  

JOHNSON: Yes, there are a few areas where AI can shine. Brainstorming is a big one. For example, I once asked ChatGPT to suggest some icebreaker games for a youth group, and it came up with a bunch of fun ideas. That kind of creative idea generation can save time when you’re planning events or looking for fresh approaches. Another area is handling technical or routine tasks. I’ve used an AI to help write bits of computer code, which not every church needs, but it shows the range of what it can do. More commonly, I’ve found it useful for drafting things like emails or letters. You can have the AI produce a first draft of a routine announcement or a thank-you note, and then you, as the human, edit and personalize it. The key is treating AI as a starting point, not a finished product. It can do the rough work, but you still refine the final output. And I’ll be honest, sometimes getting a really good result from an AI means you spend a lot of effort wording your prompt just right. At times, that effort can equal or exceed the time it would’ve taken to just write the message yourself! So it’s a bit of a trade-off. Use AI for a creative boost or quick draft but always review and correct it with your own wisdom and factual checks.  

Privacy and Security 

DAVIDSON: A lot of these AI tools are free and easy to access, which is great for churches on a budget. But you’ve raised a red flag about privacy. What do church leaders need to be aware of regarding data security when using AI services? 

JOHNSON: It’s true—“free” tools often come with hidden costs. There’s an old saying in tech: If you’re not paying for a product, then you (or your data) are the product. This absolutely applies to many free AI services. Any text, file, or information you enter into a free AI tool can be stored and used to improve the AI model. In other words, your input isn’t private. It might even resurface in some form for another user later. For example, if a pastor were to copy-paste portions of a confidential email or counseling conversation into a free AI to get suggestions, there’s a risk that sensitive information could leak out or appear in someone else’s AI results down the line. That’s a scary thought in a ministry setting, where trust and confidentiality are vital. 

Now, some paid AI plans do offer better privacy—often they promise not to use your data for training their models. But even then, it’s important to read the fine print. Not all “paid” plans are equal in this regard. The bottom line for churches is caution: Never share sensitive personal or church data with an AI tool unless you’re certain about how that data will be used and protected. It might be better to err on the side of privacy and keep confidential information out of these systems altogether or use self-hosted AI tools that keep your data in-house. In any case, church leaders should treat AI like any other online service—be wise about what you input and assume that anything you share could potentially become public. 

Anthropomorphism 

DAVIDSON: The last major concern you highlight is anthropomorphism—basically, treating these AI systems as if they were human. That’s a big word but an important idea. Could you unpack what you mean by that, and why it could be dangerous or misleading in a ministry setting? 

JOHNSON: Anthropomorphism in the context of AI means projecting human attributes onto a machine. And honestly, it’s surprisingly easy to do. I’ve even caught myself interacting with an AI chatbot like it’s a person—getting frustrated and arguing with it when it doesn’t “understand” me or asking it to brainstorm and then wondering what it “thinks” about my question. Of course, the AI isn’t thinking or feeling at all; it’s just generating responses based on algorithms. But because the conversation feels so natural, we can get a bit too comfortable, almost as if there’s a little person inside the computer. 

For the church, this is a huge issue on the horizon. If people start treating AI programs as if they have thoughts, feelings, or even spiritual significance, we could get really confused about what it means to be human. As Christians, we believe humans are uniquely created by God, made in His image, with souls and real lives that no machine can replicate. An AI might imitate conversation or even emotion, but it’s not a person. Keeping that distinction clear is going to be one of the great challenges for the church in the next few years. We’ll need to consistently teach and remind our communities about the God-given value of human life and how that differs from even the smartest machine. 

Informed, Cautious, and Hopeful

DAVIDSON: Rev. Johnson, with all these potential pitfalls—reliability issues, privacy risks, and even questions about humanity—what encouragement or advice would you give to church leaders who are trying to navigate the rise of AI? How should they approach these tools in the coming years? 

JOHNSON: My advice would be to stay informed, stay cautious, but also stay hopeful. Churches have navigated big technological shifts before by grounding themselves in Scripture and the unchanging truth of the Gospel. I believe we need to do the same with AI. Catechize your people—teach the foundational truths about God’s creation, human identity, community, privacy, and our ultimate hope in Christ. In short, approach AI with a spirit of discernment: Embrace the beneficial uses that can help your ministry but keep a watchful eye on the challenges and always lead your people back to the hope and truth we have in Jesus.  

DAVIDSON: Thank you, Rev. Johnson. This has been a fascinating discussion. Your insights give us both practical tips and a spiritual perspective, which is exactly what church leaders need as we navigate these new technologies. 

For more informative and fascinating posts like these, delivered straight to your inbox, subscribe to the CTS blog!