All micro-internships
Micro-internship brief ≈5h in one sittingno-code

AI tool showdown: 3 tools, 3 tasks, 1 verdict

Employers expect you to know which AI tool to reach for. Run the same 3 tasks through 3 different AI chatbots (9 runs total), score every run, and defend a verdict with evidence.

AI LiteracyPrompt EngineeringConversational AI

Posted by The AI Internships

The work brief

  1. 01Pick 3 different tasks: one writing, one reasoning/maths, one in Hindi/Marathi or about your local area.
  2. 02Run each task on 3 tools (e.g. ChatGPT, Gemini, Copilot, Perplexity, Claude) — same prompt each time.
  3. 03Score all 9 runs in a matrix (accuracy, usefulness, tone) and screenshot the interesting ones.
  4. 04Write a verdict: which tool for which kind of task, and why.

What you’ll produce

4 deliverables

Submission standard

Submit your 3 tasks, the full 9-run scoring matrix, screenshots as evidence, and your verdict. Make sure your screenshot shows your own logged-in account and the date, and paste a public share link to at least one run.

  • Your 3 tasks (exact prompts used)

    Written responseRequired

  • Scoring matrix: 3 tools × 3 tasks with scores and one-line notes

    Written responseRequired

  • Screenshot of the most surprising difference between tools

    Screenshot must show the tool's real interface, your logged-in account/handle, AND a visible date/time.

    Image uploadRequired

  • Your verdict: which tool for what, and why

    Written responseRequired

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 at least one of your nine runs

    A ChatGPT/Gemini/Perplexity 'Share' link to one run that also appears as a cell in your matrix.

    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.

Fair, real comparison

29%

Same prompts across tools; scores reflect actual runs, not vibes.

Evidence

14%

The screenshot genuinely shows a tool difference.

Judgement

29%

The verdict is specific and follows from the matrix.

Runs are real and first-party

29%

Auto-fail if the screenshot hides the account/date or the share link is missing or doesn't match a matrix cell (invented runs).

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.