Hey friend!
Most church leaders I talk to fall into one of two camps when it comes to AI.
Camp One: full adoption, moving fast, figuring it out as they go. Camp Two: full avoidance, waiting until someone else figures it out first. And honestly, neither camp is thinking about the right question.
The right question isn't whether to use AI. It's how to use it in a way that reflects the values your church already holds. That question matters because the tools are here now. Your staff is probably already using them. And if your church hasn't had a real conversation about guardrails, you're not avoiding the decision — you're just making it by default.
So today, I want to walk through 5 concrete guardrails your church should put in place before (or alongside) any serious AI adoption. These aren't theoretical ethics-committee talking points. They're practical decisions that protect your people, your ministry's voice, and your witness in the community.
Let's walk through each one.
Guardrail 1: AI produces drafts, not final words.
Every piece of AI-generated content that goes out under your church's name should pass through a human before it goes anywhere.
This sounds obvious until you actually watch how teams use AI in the wild. What starts as "AI writes a draft, staff edits it" quietly becomes "AI writes it, staff glances at it, and it goes out." That drift happens gradually. And it matters because your congregation didn't sign up to receive content written by a language model. They signed up to receive communication from your church.
The guardrail here is structural, not just aspirational. Build a review step into your content workflow and name a person responsible for it. That person's job is to read it, edit it, and put their fingerprints on it before it ships. One sentence added by a human makes the whole thing more yours.
Guardrail 2: Personal pastoral communication stays fully human.
There is a category of communication where AI should not touch the keyboard at all, even at the draft stage.
Care cards. Bereavement follow-ups. Notes to families in crisis. Messages to someone who just shared something vulnerable with your staff. These are the moments when people are paying closest attention to whether the care they're receiving is real. A templated response — even a well-written one — can feel hollow in those moments. An AI-written one compounds that risk.
Your team is stretched. AI can absolutely handle event announcements, bulletin copy, social captions, and dozens of other time-consuming tasks. That bandwidth recovery is real. Use it to protect more time for the communication that should never be delegated.
Guardrail 3: Data about your congregation stays inside your church.
This guardrail is the one most churches haven't thought about yet, and it's the most urgent.
When your staff pastes information about a congregation member into a public AI tool — a prayer request, a counseling summary, a giving note, details from a care visit — that information is leaving your church's environment. Depending on the platform, it may be used to train future AI models. It is almost certainly not covered by your church's data privacy practices, because most church privacy practices were written before any of this existed.
The practical decision here: define what information is off-limits for AI tools entirely. A short, clear internal policy is enough. Something like: names combined with sensitive personal information, financial data, counseling content, and health details do not go into external AI platforms. Period. That one policy closes most of the risk.
Guardrail 4: Be transparent with your congregation about how you use AI.
You don't need to disclaim every AI-assisted email. That would be overkill. But your congregation deserves to know in a general sense how your church uses these tools.
A simple paragraph in your staff policy section of the church website, a brief mention from the stage, or a note in your newsletter — any of these create the kind of low-key transparency that builds trust rather than demanding it. People are more comfortable with AI assistance than you might expect. What they're not comfortable with is feeling like something was hidden from them after the fact.
If your church ever uses AI in a way that feels significant — generating a series of teaching content, automating a care touchpoint, building a discipleship tool — proactive communication is the right call. Tell people what you built and why. That conversation usually goes better than the one you have after someone figures it out on their own.
Guardrail 5: Revisit your AI practices once a year, at minimum.
The tools available today will look different in twelve months. The risks will shift. The use cases your team discovers in month three will create new questions your original policy didn't anticipate.
A one-time conversation is not enough. Put a recurring review on the calendar — annually at minimum, quarterly if your church is using AI heavily. Bring the right people into the room: ministry staff, a tech-minded volunteer or elder, and ideally someone who pushes back on easy answers. The goal isn't to rewrite everything from scratch. The goal is to ask whether your practices still reflect your values, given what's changed.
The churches that will use AI well over the long haul aren't the ones who move fastest. They're the ones who stay curious and stay honest about what's working and what isn't.
Those are your five guardrails. Read them again slowly if you need to — each one is something your team can act on starting this week.
Now here are 2 tools to help you actually do that.
(1) The first is a word-for-word Congregation-Facing AI Transparency Statement — ready to copy, paste, and publish on your church website or drop into your next bulletin. No writing required. Just fill in your church name and contact email and it's done.
(2) The second is a 30-Minute Staff Conversation Guide — 5 questions designed to run a focused team meeting that gets everyone on the same page about AI before anyone runs ahead on their own.
Let's start with the transparency statement.
Congregation-Facing AI Transparency Statement
You do not need to overthink this. A short, honest statement published proactively does more for trust than a long explanation written after someone asks.
Here's a template you can use as-is or make your own. . .
