Knowledge & Research / Intermediate

The Fact-Checker's AI

Verify claims, trace sources, and check citations before you publish or repeat them.

8 chaptersEPUB + PDF + HTMLIntermediate guide pricingDRM-free files
The Fact-Checker's AI book cover
The problem

It starts with the situation you're actually in.

An AI-drafted competitor brief lands in your deck. One line claims a rival raised $40 million in a Series B last quarter—stated as calmly as any fact you already trust. That asymmetry is the whole problem.

The model you rely on today gets a new version number in six months. A tool you scored as trustworthy quietly changes how it sources its answers—and your old verdict no longer holds.
Outcomes

What you'll be able to do.

Find the Sentence You Don't Have Time to Check

You have eighteen flagged claims and forty minutes before a draft ships. Check them in page order and you'll burn thirty minutes confirming that "most teams use some form of project software" is true—and never reach the fabricated market-size number. You learn to check in the order that matters.

Start With the Claim at the Top of Your Queue

You learn to spot the Tier 1 claim that belongs first: high stakes, high verifiability, and usually a number—the kind your AI draft slips in as "According to the U.S." data you never actually saw.

Catch a Perfect Citation That Lies

You can dismantle a citation that looks flawless. The author is a real professor at a real university—and you still know how to confirm the source actually says what the citation claims.

Get the Process Out of One Person's Head

You can verify a claim, trace a source, and check a citation. But while those skills live only in your judgment, they walk out the door when you do. You learn to turn them into a system that stays.

Inside the book

A closer look at the work inside.

Before This Document Leaves Your Hands

Run this on the next thing you are about to send, post, or present.

  • I can point to where each specific number, date, and name came from.
  • Every quotation matches a source I have actually opened.
  • Every link resolves and supports the sentence it sits next to.
  • Nothing here is a fact I am repeating only because "an AI said so.".
  • Claims I could not verify are either cut, clearly hedged, or flagged for the reader to confirm.
  • If challenged on any single fact in this document, I have an answer.

Before You Repeat One More Thing You Didn't Check

  • Fluency masks uncertainty. The model writes a fabricated statistic in the same calm, formatted prose it uses for a verified one.
  • The errors are plausible, not absurd. A made-up funding round of $40 million is believable in a way that a made-up round of $40 billion is not.
  • Repetition launders the claim. Once a figure appears in a slide, then an email, then a board summary, it gains the authority of having been "in the deck." Nobody re-checks what.

Case Study: The Same Claim, Two Endings

You're building a market-sizing slide and ask an AI tool for a niche category's total addressable market. It returns a clean figure in a confident sentence. On the unverified path, you drop the number in. The slide looks authoritative. You see exactly where the two paths diverge.

Build Your Pattern-Spotting Reflex

Take one AI-generated paragraph you have right now—an email draft, a summary, a research answer—and run the drill. Don't verify yet; only classify. For each sentence, ask whether it carries a specific number, date, price, or ranking, and tag it NUMBER for plausible-but-wrong risk.

Spot Failure Patterns on Sight

Run this against any AI output before it earns a place in your work.

  • I have separated checkable facts from reasoning and general explanation.
  • Every specific number is tagged for verification, not accepted because it "sounds right.".
  • Every named source, study, or quote is marked "unconfirmed" until I open it.
  • Any claim about a current state, price, or "latest" version is flagged as possibly stale.
The Fact-Checker's AI visual framework
The Fact-Checker's AI frameworkInside the book
Visual preview

A diagram you can keep open while you work.

Table of contents

8 chapters, built to be read in order.

01

Trust, but Verify Everything

02

How AI Gets Facts Wrong

03

Triage Before You Trace

04

Verify the Claim

05

Trace the Source

06

Check the Citation

07

Build Your Verification System

08

Your Standing Practice

170
Pages
13,276
Words
46
Exercises, checklists & tools
8
Chapters
Formats

Three formats. One purchase.

EPUB, PDF, and HTML are included so the book can work on an e-reader, as a designed copy, or as a searchable desk reference.

EPUB

For e-readers and reading apps.

PDF

The designed edition with diagrams and layouts intact.

HTML

Searchable, copy-pasteable, and practical as a reference.

Complete guide

The Fact-Checker's AI

$9.99
Intermediate guide pricing
  • EPUB + PDF + HTML included in one purchase
  • 8 chapters from the complete guide
  • DRM-free files for your own devices
  • 7-day refund review for duplicate purchases, access issues, wrong files, or materially defective downloads
Add to cart - $9.99
Secure checkout / instant download / tax handled at checkout
Before you buy

Questions, answered.

Does this include the full book?

Yes. You get the complete edition, including the chapter sequence and internal materials described on this page.

Which formats are included?

EPUB, PDF, and HTML are included so you can read on an e-reader, keep a designed copy, or use the searchable browser version.

What is the refund policy?

Because this is an instant digital download, broad change-of-mind refunds are not offered after the files have been accessed. Refund requests are reviewed within 7 days for duplicate purchases, accidental purchases before access, access failures we cannot fix, wrong files, corrupted files, or pages that materially misdescribe the book.