Should Your Church Have an AI Policy? What the Data Says

AI policy for churches 2026 conversations have picked up fast, and the numbers explain why. A new report found that 64 percent of church leaders believe it's important to have an AI policy in place.

Only 5 percent actually have one written down. That gap is bigger than almost any other technology issue churches are facing right now.

This isn't a hypothetical concern anymore. Staff and volunteers across the country are already using AI tools for sermon prep, social media captions, email drafts, and bulletin copy, often without anyone at the church having said yes or no to it officially.

If your church hasn't talked about this yet, you're already behind most of the conversation, but you're also in good company, since almost everyone else is behind too.

What Does the 2026 Data Actually Show?

The clearest picture comes from the 2026 State of Church Technology report, developed by Pushpay in partnership with Barna Group, which surveyed more than 1,300 church leaders.

The report found that AI adoption in churches has moved past the "should we use this" stage. Leaders are no longer debating whether AI belongs in ministry work. The bigger, harder question now is how to use it responsibly. A separate 2026 survey covered by ChurchTechToday put a sharper point on this, finding that 93 percent of pastors are already using AI in some form, most commonly for sermon prep, content creation, and administrative tasks.

At the same time, concern hasn't gone away. Church leaders surveyed pointed to plagiarism, loss of authenticity in preaching, message integrity, and data privacy as their top worries. Roughly 83 percent expressed at least some concern about privacy and security specifically.

Put those two numbers together, high usage and high concern, and you get exactly what's driving the policy gap. People are using the tools faster than anyone is setting rules around them.

Why Does This Matter for a Church Specifically?

A church holds a different kind of information than most small businesses. Prayer requests, counseling notes, giving records, and membership details all carry a level of trust that goes beyond typical customer data.

StartCHURCH's guide on responsible AI use points out that even well-intentioned use of an AI tool could expose private information if sensitive data gets uploaded into a system that wasn't built to protect it. 

That's the real risk behind the policy gap. It's rarely about a staff member doing something reckless on purpose. It's about nobody having defined what's off-limits in the first place.

This is the same reasoning behind the AVL planning conversations more churches are having lately. We touched on this in What Is Church AVL Integration? All You Need to Know. Where the core idea is that technology decisions work best when they're planned around the whole church's mission. Not adopted piece by piece without a clear framework.

What Should Actually Be in a Church AI Policy?

A good church AI policy doesn't need to be long or complicated. Subsplash's 2026 guide on building church AI policies recommends starting with a simple, non-negotiable rule: never upload personally identifiable information, like member names, giving records, or private prayer requests, into a public AI tool.

Beyond that starting point, most workable policies cover a few consistent areas:

• A purpose statement: Why is the church using AI at all, and how does that connect back to its mission? This keeps the policy grounded instead of feeling like a generic tech document.
• Approved use cases: Sermon outlining, social captions, email drafts, and bulletin copy are common low-risk starting points many churches approve first.
• Prohibited uses: This is where member data, counseling notes, financial records, and anything tied to a real person's private information are explicitly ruled out.
• A review process for new tools: Before any new AI tool gets adopted church-wide, someone should check its data privacy terms and confirm it won't train on or store church data without permission.
• Disclosure expectations: Some churches require a simple internal flag when a piece of content, like a small group guide or a public statement, is substantially AI-generated so a human reviews it before it goes out.
• A review schedule: AI tools and their terms of service change quickly. A policy that gets revisited annually stays useful. One that's written once and forgotten usually doesn't.

None of this requires a lawyer or a six-month committee process. A single page that your staff and volunteers actually read is worth more than a lengthy document nobody opens.

Does Having a Policy Mean Restricting AI Use?

Not necessarily. A policy can just as easily open the door to responsible use as it can restrict it.

Some churches choose to limit AI significantly, especially around anything touching sermon writing or pastoral care. Others encourage it for administrative efficiency, freeing up staff time for the parts of ministry that genuinely need a human. 

Both approaches are valid. The point of the policy isn't to pick a side for every church everywhere. It's to make sure your specific church has actually made a decision, instead of leaving it up to whoever happens to be drafting the next newsletter.

This mirrors a pattern we see constantly in production planning too. Churches that plan their technology intentionally, rather than adding pieces reactively, tend to end up with systems their teams can actually manage. 

If your church is working through that kind of planning on the production side, How Small Churches Can Improve Worship Production Without a Big Budget walks through a similar mindset applied to equipment and staffing decisions.

What Happens If a Church Skips This?

Staff and volunteers will keep using AI tools regardless of whether a policy exists. The Subsplash research found that 100 percent of surveyed church communicators expressed interest in using AI for content creation, even among churches that hadn't formally addressed it yet.

Skipping a policy doesn't stop AI use. It just means the church has no say in how it happens. That's the real cost of the 64-to-5 gap. It's not that churches are unaware AI matters. It's that awareness hasn't turned into action yet.

Summing It Up

The 2026 data is clear on two things. Church leaders overwhelmingly believe an AI policy for churches matters, and almost none of them have actually written one. 

That gap represents real risk, mainly around data privacy and message integrity, but it also represents an easy win. A short, values-driven policy covering approved uses, prohibited uses, and a basic review process closes most of that gap in an afternoon, not a quarter.

Technology decisions, whether it's AI, your production system, or how your church streams a service, work best when they're made on purpose instead of by accident.

Want to talk through how your church's technology decisions fit together, from AI tools to your production setup? Book a free consultation with our team.

Read Also:

• What Is Church AVL Integration? All You Need to Know
• How Small Churches Can Improve Worship Production Without a Big Budget
• Explore our Video services

Frequently Asked Questions

• Does a small church really need an AI policy, or is this just for large churches? 

Size doesn't change the risk. The 2026 AI in Church survey found that the majority of respondents lead congregations under 500 people, which confirms AI adoption isn't limited to large, well-resourced churches. A one-page policy takes an afternoon to draft and protects a small church just as much as a large one.

• What's the biggest risk of not having an AI policy? 

Data privacy is the most commonly cited concern, with roughly 83 percent of church leaders expressing at least some worry about it. The specific risk is a staff member or volunteer uploading sensitive information, like member records or prayer requests, into a public AI tool that wasn't built to protect it.

• Should a church ban AI outright instead of writing a policy? 

Some churches choose to restrict AI heavily, and that's a legitimate choice. But even a church that wants to limit AI use benefits from writing that decision down, since staff and volunteers may already be using AI tools without realizing there's an expectation not to.

• How often should a church update its AI policy? 

Most guidance points to reviewing it at least once a year. AI tools and their data practices change quickly, and a policy that isn't revisited tends to fall out of date fast.