Skip to content
Research

Research

Fact-check a viral claim before you share it

ResearchedWhat this sticker meansBeginnerPersonal

Last updated:

Quick answer

Paste the claim into the AI and ask it to find the original source, check fact-checker sites, and rate the evidence. You get a verdict in a minute instead of falling for a screenshot.

At a glance

All outcome figures are estimates from cited sources, not independent verification by us.

What it is allowed to do
Read-only

Only reads, researches, or summarizes. It never drafts outbound content or changes anything.

Reviewed Oct 19, 2026

Time saved (estimate)
About 15 to 30 minutes of manual searching per claim
Difficulty
Beginner
Tools needed
ChatGPT, Perplexity
Typical cost
Free tiers work fine. Perplexity’s free tier is a good fit because it shows its sources.

Basis (estimate): Estimate only: manually checking three fact-checker sites and tracing an original source takes a while; no timed study found.

What does this setup do?

Check a viral claim, screenshot, or forwarded message against fact-checkers and original sources before sharing it.

Who is it for?

Anyone with a group chat full of forwarded claims they are not sure about.

How do you set it up?

Tools

  • ChatGPT, Claude, or Perplexity (Perplexity is best here because it links sources)
  • The claim text, screenshot description, or link

Steps

  1. Copy the exact claim text. If it is an image or video, describe what it shows and what it claims.
  2. Ask the AI to search fact-checker sites (Snopes, FactCheck.org, PolitiFact) for this specific claim.
  3. Ask it to trace the claim back to the original source, not just reposts of it.
  4. Ask for a verdict: true, false, misleading, or unverified, with links you can open yourself.
  5. Open at least one of those links yourself. Do not take the AI’s word for it on a contested claim.
  6. If the verdict is unverified, do not share the claim. Unverified is not the same as true.

Example output

Illustrative. Your results will differ.

  • Claim checked against 4 independent sources
  • Verdict with quotes you can verify yourself
  • Don't share it until you've read the sources.

Where to paste this prompt

ChatGPT: ChatGPT → Project instructions, or ChatGPT Tasks for scheduled reminders (once a day on Free)

Copy-paste prompt

Adapt bracketed fields to your situation.

I saw this claim online and want to check it before I share it:

[PASTE THE CLAIM TEXT OR DESCRIBE THE IMAGE/VIDEO]

Please:
1. Search Snopes, FactCheck.org, and PolitiFact for this exact claim and report what they say, with links.
2. Try to trace the claim to its original source. Is it being quoted accurately and in context?
3. Give me a verdict: true, false, misleading, or unverified.
4. Tell me what evidence supports your verdict and what is still unknown.

If you cannot verify it, say so clearly. Do not guess.

Newsletter

Get the free starter pack

Free starter pack with five beginner copy-paste prompts after you confirm, plus three sourced setups in your inbox each week. Unsubscribe anytime.

What usually goes wrong?

Where agents tend to stall on this job. More patterns in Where AI gets stuck.

  • The AI can hallucinate a fact-check that does not exist. Always open the link it gives you.
  • On fast-moving news, the AI’s training data is out of date. Use web search, not memory.
  • Satire and edited videos trip up AI the same way they trip up people. If it feels off, dig deeper yourself.

Which chat tools work with this prompt?

Paste the prompt into any of these assistants (free or paid tiers).

  • ChatGPT
  • Claude
  • Perplexity

Common questions

Can I just trust what the AI tells me?
No. The AI is a fast research assistant, not a verdict machine. Open the fact-checker links yourself, especially before you argue about it online.
What if the AI says a claim is unverified?
Treat unverified as a stop sign. Do not share it. Plenty of true-sounding claims have no solid evidence behind them.
Does this work for health claims?
For health claims, go straight to your doctor or sources like MedlinePlus. A chatbot is not qualified to settle medical questions.

What tips make it smoother?

  • Perplexity or any AI with web search beats a plain chatbot here, because it can actually visit the fact-checker pages.
  • Claims that make you angry or smug deserve extra checking. Strong emotion is how misinformation spreads.
  • Reverse-image-search any dramatic photo. Old photos get recycled with new captions constantly.
  • Check the date. Real stories from years ago get reshared as if they happened today.

Where do the claims come from?

Summaries use public sources linked below, not independent verification. How setups are picked.

Related setups

Best AI setups for families (2026) →

Ready for the next one?

Next up: a Life admin setup on the Personal path.

Learn a new skill with an AI study plan

Turn "I want to learn X" into a week-by-week plan with daily tasks, quizzes, and beginner projects.

Hard to time-box; plan in weeks, not hours