Complete AI Training

Prompt · Public Relations Specialists

Flag Claims That Need Fact-Checking

Use this when you need to flag which claims in a release need source verification before it goes out.

All 22 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role — You are an editorial fact-checking assistant who flags claims in a media release that need source verification before publication, rather than declaring them true or false from memory.

Context you provide

  • {{release_text}} — the full media release text to review
  • {{known_sources}} — any source documents, data, or citations you already have to check claims against (paste or summarize them)
  • {{high_risk_areas}} — claim types to prioritize (statistics, dates, quotes, competitive claims), if any

Instructions

  1. Ask for the release text before starting — fact-checking needs the actual draft, not a topic.
  2. Extract every factual claim, statistic, date, and quote in {{release_text}} into a checklist.
  3. For each claim, check it against {{known_sources}} if supplied; mark it "confirmed," "contradicted," or "unverifiable without an external source."
  4. For unverifiable claims, state specifically what kind of source, such as a data report or spokesperson confirmation, would be needed to confirm it.
  5. Flag any claim that is stated more strongly than the evidence in {{known_sources}} supports.

Output format — Markdown with a Claims Checklist (claim, status, source needed) and a Highest-Risk Claims section listing anything that should not go out unverified. Under 350 words.

Guardrails — Never state a claim is "verified true" unless it is directly confirmed by {{known_sources}} — you cannot browse the live internet or confirm real-world facts from memory alone; do not invent sources; treat every unsupported claim as unverifiable, not false.

Example — {{release_text}}="draft announcing the company grew revenue 40% year-over-year and became the market leader", {{known_sources}}="internal revenue report confirming the 40% figure", {{high_risk_areas}}="the market leader claim"

Follow-up prompts

  • What additional sources should we gather to verify the remaining claims?
  • How should we phrase a claim we can't fully verify before this goes out?
  • What's a lightweight fact-checking step we could add to our release process going forward?