Use AI to Polish, Not to Preach

Should Preachers Use AI to Write Sermons? A Practical Take for Writers

Short answer: yes, but with guardrails. Most writing tools already use AI to correct grammar, suggest phrasing, and reference sources faster. That's an assist, not a replacement.

The real question is whether a machine should draft core content that carries faith, conviction, and local nuance. That's where the line appears.

What AI is good at (and where it falls short)

Scott Hozee of the Center for Excellence in Preaching put it plainly: "If a sermon is just information, AI will do. But if a sermon is also about passion and inspiration and faith and the personality and gifting of a specific person who has been called to be a preacher, then AI cannot do any of that."

On a Church Juice podcast, Kenny Jahng offered a useful frame: treat chatbots like "a very very green seminary student." Helpful. Not in charge.

Reality check: risks and limits

• Accuracy is inconsistent. Confident, polished text can still be wrong.
• Source issues exist. Much training data was scraped without explicit permission, and litigation is ongoing. See the U.S. Copyright Office's guidance on AI and copyright for context: copyright.gov/ai.
• Compute costs are huge. Don't assume the tech's current pace equals maturity.
• Cognitive offloading is real. Outsource too much and your creative edge dulls.

Use AI for the grunt work, keep the soul human

• Brainstorming: ask for 10 outline options, angle variations, or question prompts to pressure-test the main idea.
• Research scaffolding: request a list of themes, cross-references, or historical context you can verify with trusted sources.
• Language polish: tighten sentences, fix grammar, and reduce filler.
• Illustration seeds: generate metaphor ideas and story prompts, then replace with your own lived examples.
• Logistics: condense draft sections to slide bullets or a speaking outline.

Do not outsource these

• Exegesis and theological judgment.
• Pastoral sensitivity and local context.
• Personal testimony and lived experience.
• Prayerful discernment of tone, call, and timing.
• Final structure and throughline.

Quality control checklist for writers

• Ask AI for sources, then verify with trusted commentaries and primary texts.
• Fact-check names, dates, quotes, and Scripture references.
• Run a plagiarism check on any AI-generated passages you keep.
• Rewrite in your own voice. Read it aloud. If it doesn't sound like you, it isn't you.
• Audit originality: include one personal story, one local example, and one clear takeaway.
• Document sources you cite or paraphrase and secure permissions when needed.

A simple sermon-writing workflow with AI (45-90 minutes)

• Clarify the thesis: one sentence that ties Scripture to a specific need.
• Prompt AI for 3 outlines with different structures (problem-truth-application, story-exegesis-call, etc.).
• Merge the best parts. Fill in exegesis and application yourself.
• Prompt for 10 illustration ideas. Replace most with your own stories.
• Polish: ask AI to suggest cuts for clarity and flow. Keep what fits your voice.
• Verify all references. Add citations. Read aloud and adjust pacing.

Ethics and attribution

• Be transparent with your team about AI's role.
• Never paste private pastoral data into a public model.
• Cite any direct quotes and check licenses for external material.

The bottom line

Use AI like an intern: fast, helpful, and under close supervision. You own the message. You carry the Spirit, the context, and the courage to say it in the room.

Let the machine propose. Let the preacher decide.

Further learning

• AI and copyright policy overview: U.S. Copyright Office
• Practical training and workflows for writers using AI: Complete AI Training - Courses by Job