Sermon Prep
Could This Report Be Quietly Feeding Me Something Theologically Off?
Work through every fear this series has covered — invented facts, ghostwriting, weak review — and one question still sits underneath all of it: what if a cited scholar holds views you’d never knowingly build a sermon on? What if the AI can’t tell orthodoxy from heterodoxy and just blends whatever it finds? That’s the deepest version of the worry in this series, and it deserves the most direct answer.
This isn’t a worry unique to one tool
Korean church bodies have already staked out public ground on exactly this. The Presbyterian Church of Korea (Tonghap), at its 109th General Assembly, adopted an “AI-Era Pastoral Ethics Declaration,” prompted by growing use of tools like ChatGPT for sermon research and prayer-writing and a felt urgency to set ethical boundaries around AI use in church and society. Chongshin University’s 2025 generative-AI guidelines, mentioned earlier in this series, work from the premise that “AI, as a tool used by humans made in God’s image, must remain under human ethical control and guidance,” and restrict its use accordingly. Both land in the same place: not a blanket rejection of AI, but controlled use inside theological boundaries.
How the boundary actually holds, structurally
Didymus Lab enforces that boundary on two layers.
First, the source pool itself is restricted. As covered earlier in this series, material is limited to openly and verifiably licensed scholarship — CC-BY, CC0, and public domain. That’s an accuracy safeguard, but it’s a doctrinal one too: sources like the SBL Greek New Testament, the Westminster Leningrad Codex, peer-reviewed journal articles, and patristic writings simply aren’t a channel through which unverified claims circulating online can enter in the first place.
Second, content itself is gated against orthodox doctrinal boundaries. Conclusions built on religious pluralism, syncretistic readings that blend other religious traditions in, or positions that contradict historic Protestant sexual ethics are designed not to surface in a report at all. This isn’t after-the-fact censorship — it’s upstream design. The goal is that the material-gathering stage itself can’t arrive at those conclusions to begin with.
From fear to verification
As this series opened by saying: your caution is warranted. What matters is following that caution all the way through — not stopping at “should I use this at all,” but asking what it’s built on, what boundaries constrain it, and whose final review it passes through. Answering that honestly is our job. Open a sample report and check the footnotes and sources yourself, and if a question remains, reach us any time at didymus@didymuslab.com.
Thank you for reading this series through. The fear may not vanish entirely — but we hope it’s now clearer exactly what to check.
Comments
Leave a comment
Comments are reviewed before being published. Only your name and comment are stored — no personal data collected.
You might also like
Sermon Prep
If AI Writes the Sermon, Is It Still Mine?
The deepest objection to AI sermon prep isn't laziness — it's identity. A prominent evangelical voice made that case publicly. We take the critique seriously, then draw the line that actually matters: between gathering material and deciding what to preach.
Sermon Prep
Afraid an AI Will Make Something Up and You'll Preach It?
The biggest fear around generative AI isn't laziness — it's a confident lie. Here's why AI 'hallucination' happens, why it's a reasonable thing to worry about, and how verified sources plus automated checking gates are built to catch it before it reaches a pulpit.
Sermon Prep
The Hours You Used to Spend in Commentaries — Where Do They Go Now?
AI's real value in sermon prep isn't finding more sources — it's the harder job of narrowing them down. Here's why more material usually means more prep time, not less, and why selection, not search, is the thing actually worth automating.
Loading comments…