Project 6 — The Discernment Audit

“But test everything; hold fast what is good.” — 1 Thessalonians 5:21 (ESV)

Chapter: 6 — Deepfakes, Bots, and the Digital Mission Field Deliverables: discernment-log.txt and audit-reflection.docx Tools: A browser and a real Bible (esv.org or a free Bible app) — see Appendix A. A free AI assistant is optional and, if used, gets verified and noted.

The brief

This week you learned that in the digital mission field, verification is the ministry — because an apologist whose sources don’t check out has already spent the credibility that belongs to Christ’s message. Reading that is one thing. Living it, against your own real feeds, is another.

So this project makes it real. You are going to become, for one week, the person in your circles who checks first. You’ll take five faith-related claims that are actually circulating right now — in your feeds, your family group chat, your church’s social pages — and run each one all the way to a verdict, showing your work at every step. Then you’ll do the honest thing almost no one does: name the one you almost shared before you caught it.

Real claims only. No inventing convenient examples. The value of this audit is that it trains the reflex on the exact kind of content that actually reaches you.

What you’ll build

Two files:

1. discernment-log.txt — a plain-text log of five real claims, each run through the 30-second check with every step recorded. This is your evidence: not just verdicts, but the trail that produced them.
2. audit-reflection.docx — a one-page reflection (roughly 400–500 words) on what the audit taught you, built around the item you almost shared.

Steps

1. Collect five real claims. Over a few days, screenshot or copy five faith-related items circulating in your own feeds — quotes attributed to Christians, “news” about the Bible or archaeology, shocking screenshots, dramatic conversion or persecution stories, AI-written history. Aim for variety, and include at least one you initially believed.
2. Log each through the three questions. In discernment-log.txt, give every claim its own block. Record: the claim in one line; who actually said it (traceable source or not); where’s the original (the real home, or “only meme graphics”); who else reports it (independent corroboration, or none). Note any reverse-image or exact-phrase search you ran.
3. Reach a verdict for each. End every block with one of three: True — safe to share, Needs context — misleading as posted, or False — don’t share — plus one sentence of reasoning. Be willing to land on “true”: discernment isn’t reflexive cynicism.
4. Flag the near-miss. In your log, mark the one claim you were closest to sharing before you checked it. This is the heart of the reflection.
5. Write audit-reflection.docx. Tell the story of the near-miss honestly — what it was, why it got past your guard (be specific: did it flatter your team? confirm a fear? come from someone you trust?), and what tipped you off. Then draw one lesson you’ll carry into your feed, and connect it to the week’s Scripture (1 Thess 5:21, Prov 14:15, or 1 John 4:1). If you used an AI assistant anywhere in the audit, add a one-line AI usage note: what you asked, what it got wrong, what you verified.

What good looks like

A strong audit reads like a paper trail, not a set of opinions. I can see how you knew, not just what you concluded — the phrase you searched, the second outlet you looked for, the original you found or failed to find. Your five verdicts aren’t all “false” (that would mean you cherry-picked); at least one survives scrutiny and you say why. And your reflection is genuinely honest about the near-miss — the strongest submissions admit that the claim slipped past precisely because the author wanted it to be true. That self-awareness, not a perfect scorecard, is the skill this week is building.

Grading rubric — Normal (out of 100)