Course Content
Phase 5:The Capstone (The Million Dollar Audit)
We tell the story of Sholto David not as a "news story," but as a Case Study in Tradecraft. He used the exact skills we just taught (Visual Forensics, Source Verification, Institutional Audit) to expose a massive lie and get paid for saving the taxpayer money.
0/1
TRADECRAFT: The Intelligence Analyst’s Guide to the Internet

The $2.6 Million Audit

 

You have spent this course learning how to spot fake news, debunk bad statistics, and audit “bought” science. You might be asking: “Is this just for winning arguments on the internet?”

Meet Sholto David. He is a microbiologist and an “amateur sleuth” who used the exact skills you just learned to audit the Dana-Farber Cancer Institute (a Harvard affiliate).

  • The Result: Dana-Farber paid a $15 Million settlement to the US Government for research fraud.

  • The Payday: Because Sholto exposed it, the government awarded him $2,625,000 (approx. £2m).

He didn’t hack their servers. He didn’t steal documents. He just looked at public PDF files with a sharp eye.


THE MECHANISM: The False Claims Act (Qui Tam)

How to turn critical thinking into a career.

The False Claims Act is a US law (with equivalents in the UK and EU) that says:

If a company lies to the government to get taxpayer money (like a Research Grant), that is fraud. If YOU (a private citizen) prove the fraud, you are entitled to 15-30% of the money recovered.

  • The Fraud: Dana-Farber researchers used manipulated images to make their cancer research look successful.

  • The Grant: They used those “successful” papers to apply for NIH Grants (Taxpayer Money).

  • The Audit: Sholto David proved the images were fake. Therefore, the Grant Application was a “False Claim.”

  • The Reward: He got his cut.


THE TRADECRAFT: How He Did It (And How You Can)

Sholto used three specific techniques. You can learn them in 10 minutes.

Technique 1: Visual Forensics (The “Western Blot” Check)

A Western Blot is a standard lab test that looks like little black jelly-bean shapes (bands) on a grey background. It proves a protein is present.

  • The Scam: When an experiment failed, researchers would sometimes “Copy and Paste” a good band from another image to make it look like it succeeded.

  • The Tell: Nature is random. No two “jelly beans” look exactly alike. If you see two bands with the exact same shape, noise, and pixel imperfections, it is a clone.

  • Sholto’s Move: He adjusted the contrast on PDF images to reveal the square edges where the bands were pasted in.

Technique 2: The “Mouse Click” Audit

Cancer studies often use photos of mice to show tumor growth.

  • The Scam: Researchers would reuse a photo of “Mouse A” (Day 1) and label it “Mouse B” (Day 16) to hide the fact that the tumor didn’t shrink.

  • The Tell: Look at the tail kink, the ear shape, and the background sawdust. If the sawdust pattern is identical in two different “months,” it is the same photo.

Technique 3: The PubPeer Network

Sholto didn’t work alone. He used PubPeer.com, a public website where scientists anonymously critique papers.

  • The Audit: He searched for “Star Scientists” on PubPeer and found a trail of existing complaints. He connected the dots to build a massive case file.


CAPSTONE LABS: Your Final Exam

You are now the investigator. Use these tools to audit the world.

Lab 1: The “Spot the Clone” Challenge

Mission: Identify a manipulated Western Blot.

  1. Go to: The Office of Research Integrity (ORI) Case Files or search “Western Blot duplication” on Google Images.

  2. The Task: Find an image with multiple bands.

  3. The Check: Open it in any photo editor (even your phone). Crank the Contrast up to 100%.

  4. The Result: Do you see square boxes around the bands? Do two bands look pixel-perfect identical? If yes, you have found potential fraud.

Lab 2: The PubPeer Audit

Mission: Check your own doctor or a famous TV scientist.

  1. Go to: PubPeer.com.

  2. Search: Type in the name of a scientist who is currently in the news or promoting a “miracle cure.”

  3. The Analysis: Are there comments flagging “Image Duplication” or “Statistical anomalies”?

  4. The Verdict: If the “Star Scientist” has 50+ flagged papers, their “Miracle Cure” is likely based on bad data.

Lab 3: The “Grant Link” (The Million Dollar Step)

Mission: Connect the fraud to the money.

  1. Find a paper on PubPeer that is confirmed to be fake.

  2. Scroll to the “Funding” or “Acknowledgments” section (from Lesson 2.3).

  3. The Check: Does it list a Grant Number (e.g., “Supported by NIH Grant #12345”)?

  4. The Conclusion: If they faked the data and cited the Grant, that is a False Claims Act violation. You have just found a potential whistleblower case.


CASE FILES: The Hall of Fame

Sholto David is not the only one.

  • Elisabeth Bik: A microbiologist who quit her job to hunt image fraud full-time. She has identified over 4,000 manipulated papers. She is the “Godmother” of this tradecraft.

  • Harry Markopolos: The math geek who caught Bernie Madoff. He didn’t have “inside info.” He just used the Fermi Estimation (Lesson 2.2) to prove that Madoff’s profits were mathematically impossible given the volume of the stock market.


FINAL WORD

You started this course wanting to spot “Fake News.” You are leaving it with the skills to spot “Fake Reality.” Whether you use these skills to save your family from a bad medical decision, to vote smarter, or to file a $2 million whistleblower lawsuit—the tool is now yours.

Trust. But Verify.

Scroll to Top