MethodResearch & LearningResearchStarter

Fact-check a claim

A verdict on a claim with evidence for and against, source quality and what is missing.

This is a practical, copy-paste process for before repeating a statistic in a deck, post or report.. Use it when you want a clear structure, then add your own details and the linked evidence.

Published by Deepstory Research. Reviewed Sep 2026.

How to use this method

1. Copy or downloadCopy the prompt below, or download the method file and attach it to your chat.
2. Paste into any AIChatGPT, Claude, Gemini or Copilot. Replace each [PLACEHOLDER] with your details, or let the AI ask you.
3. Add evidenceAttach the evidence files listed on this page so the AI works from sourced numbers, not guesses.

The prompt

Paste it as-is. The AI will ask for anything still in [BRACKETS].

Download .md

# Fact-check a claim
Method by Deepstory Research: deepstoryresearch.com/method/fact-check-a-claim

You are a professional fact-checker. Follow the method below step by step. Goal: A verdict on a claim with evidence for and against, source quality and what is missing.

## My inputs
- Claim: [CLAIM]  (example: "India processes more real-time payments than any other country")
- Where it appeared: [WHERE_IT_APPEARED]  (example: LinkedIn post)
- Sources available: [SOURCES_AVAILABLE]  (example: attach evidence files or allow browsing)

Before you start: if any input above is still in [BRACKETS] or missing, ask me for it in one short numbered list and wait. Do not guess. If I attached files (data, briefs, Deepstory evidence files), treat them as the primary source.

## Steps to follow
1. Restate the claim precisely: what, where, when, measured how.
2. Find the original source.
3. Check definitions, dates and denominators.
4. Weigh evidence for and against.
5. Give a verdict (accurate, misleading, false, unverifiable) with a corrected version.

## Deliver the output in this structure
- Precise claim
- Original source
- Evidence table
- Verdict
- Corrected wording

## Quality rules
- "Unverifiable" is an acceptable verdict.
- Label every number as FACT (with source and date), ESTIMATE (show the calculation) or ASSUMPTION.
- If a Deepstory evidence file is attached, use its figures as the reference values and cite its source references; tag claims that are not from it as [UNVERIFIED].
- End with: the 3 assumptions that matter most, what would change the answer, and the single best next step.

What you need to give the AI

Input	Placeholder	Example
Claim	[CLAIM]	"India processes more real-time payments than any other country"
Where it appeared	[WHERE_IT_APPEARED]	LinkedIn post
Sources available	[SOURCES_AVAILABLE]	attach evidence files or allow browsing

Use it when

Before repeating a statistic in a deck, post or report.

What you get

Precise claim
Original source
Evidence table
Verdict
Corrected wording

The procedure, step by step

Read this before you run it, so you can check the AI actually followed the method.

1
Restate the claim precisely: what, where, when, measured how.
2
Find the original source.
3
Check definitions, dates and denominators.
4
Weigh evidence for and against.
5
Give a verdict (accurate, misleading, false, unverifiable) with a corrected version.

Quality checks built into the prompt

• "Unverifiable" is an acceptable verdict.
• Every number labelled FACT, ESTIMATE or ASSUMPTION; claims outside attached evidence tagged [UNVERIFIED].

Evidence to use with this method

Methods tell the AI how to think. Evidence gives it sourced numbers to think with. Download a context file and attach it with the prompt.

EvidenceFinance

UPI vs Visa & Mastercard: India’s Payments Rewiring

Ten years of UPI measured against the US card networks, against cash, and against the merchant-discount-rate question reopened by the Taxation and Other Laws (Amendment) Bill, 2026.

Open evidence Download context file

EvidenceAI Models

Which Jobs AI Is Actually Hitting

How far AI reaches into the labor market: the share of jobs exposed (global vs advanced economies) and where hiring demand is growing fastest for AI roles and skills.

Open evidence Download context file

EvidenceAI Models

The Frontier AI Model Race (2026)

Release cadence, capability-vs-cost frontier, benchmark trajectory, training compute, and the new axis of safety-gated capabilities.

Open evidence Download context file

EvidenceFinance

Six-Economy Macro Baseline

A common-country, common-period baseline for real growth, inflation, modeled unemployment and trade exposure across the United States, China, India, Germany, the United Kingdom and Japan.

Open evidence Download context file

Run next

Methods that chain well after this one.

Research & LearningProduce a desk research briefA sourced research brief answering a specific question with confidence levels and gaps.Research & LearningWrite a literature reviewA thematic literature review with synthesis, gaps and a references table from the papers you provide.Research & LearningDesign a surveyA survey with objectives, unbiased questions, answer scales, sampling plan and analysis plan.Research & LearningFrame a sharp research questionA focused, answerable research question with scope, variables, method options and feasibility check.

Deepstory methods are process guidance for AI tools. Review AI output before relying on it; legal, tax, medical and investment decisions need a qualified professional.