Two posters, five people, three seconds each
A poster earns about 3 seconds of attention. Generate 2 genuinely different posters for the same event with AI + your own editing, show each to 5 real people for 3 seconds, and declare a winner based on data, not taste.
Posted by The AI Internships
The work brief
- 01Pick a real (or realistic) event and design 2 posters with different strategies — e.g. image-led vs. type-led — generating options with an AI image tool and refining each by hand.
- 02Run the test: show each person ONE poster for 3 seconds, then ask — what’s the event? when? would you go? (Alternate which poster each person sees.)
- 03Record every answer verbatim: 5 people minimum, at least 2 per poster.
- 04Score which poster transmitted the key facts better, and diagnose why.
What you’ll produce
Submission standard
Submit both posters in one image, the raw test answers, and your data-based verdict, plus your AI workflow. Also submit the public Canva view-link to your posters and a screenshot of your own editor showing your account name and today's date.
Both posters, side by side, labelled A and B
Image uploadRequiredThe raw test: each person, which poster, their verbatim answers
Log all 5 people, which poster each saw, and their verbatim answers — real people misremember and contradict; flat, tidy, or invented answers fail, and the numbers must reconcile with your verdict.
Written responseRequiredThe winner, by the numbers — and the design reason you think explains it
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 Canva view-link to your two posters
Canva 'Share > Anyone with the link can view'. Must open to YOUR editable posters — a dead or private link fails.
Public linkRequiredScreenshot of your OWN editor building a poster — account name and today's date visible
Show the live canvas with your username and system date, not just the exported posters.
Image uploadRequired
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.
Two real strategies
14%The posters differ in strategy, not decoration.
Real recall test
29%The answers read like real people misremembering things — messy and specific.
Data-led verdict
29%The winner follows from the answers, with a sound design diagnosis.
First-party provenance
29%Auto-fail if the posters have no live editable file, if the editor screenshot lacks account/date, or if the recall answers read invented/flat or do not reconcile with the verdict's numbers.
How we grade your AI usage
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.