All micro-internships
Micro-internship brief ≈3h in one sittingno-code

Catch the AI lying

AI sounds confident even when it’s wrong. Ask an AI a factual question, find where it hallucinated, correct it with a real source — then confront it and see if it backs down.

AI LiteracyCritical AI Judgment

Posted by The AI Internships

The work brief

  1. 01Ask an AI a factual question in a topic you know.
  2. 02Find at least one claim that is wrong or unverifiable.
  3. 03Correct it and back the correction with a credible public link.
  4. 04Confront the AI with your source and record whether it corrects itself or doubles down.

What you’ll produce

3 deliverables

Submission standard

Paste the AI’s claim, the truth, and a public source link we can check. Include your AI workflow — your prompts and how the AI reacted when you confronted it. Add a public share link to the real chat, including where you pushed back with your source.

  • The AI’s claim you’re challenging

    Written responseRequired

  • The correct answer, in your words

    Written responseRequired

  • A public source link that backs your correction

    Public linkRequired

You’ll complete these inside your private workspace.

What you must submit as proof

This brief requires evidence an AI can’t fabricate.

  • Public share link to the actual AI chat

    Must be YOUR live chat containing the original claim AND the moment you confronted it with your source.

    Public linkRequired

Submissions without this evidence cannot be submitted.

Protect other people in your proof. Blur faces, names, phone numbers and email addresses before you upload, and refer to anyone you worked with by role or number ("Listener 1", "the stall owner"). Your proof is only ever used to check your work — it is never published, never appears on your certificate, and is never shown in your public portfolio.

How your work is evaluated

The passing benchmark is 70/100.

Correctness

33%

The correction is actually right.

Sourcing

33%

The linked source genuinely supports the correction.

Verifiable AI exchange

33%

Auto-fail if the chat link is missing/dead or doesn't actually contain the claim and the confront-and-respond moment (invented exchange).

How we grade your AI usage

30% of your score

Using AI is the point — it’s the skill this certificate proves. You’ll answer three short questions about how you used it: what you asked, what was wrong with its first answer, and what you changed. Specific, honest answers score high. “I pasted the brief and submitted the answer” scores near zero.