Skip to main content

MCQ creation

Feature RequestPractice Test

AI HAS to be able to create better confounders for MCQs.

AI should be able to read the question, understand the context and create confounding questions that are SENSIBLE (isn’t that literally the job of AI?) and of prescribed difficulty. If I ask “How many hands do humans have?”, the MCQ selections shouldn’t be: “1. Blue 2. Two 3. Car 4. Up.” You’re telling me a script doesn’t exist that can direct the AI engine to actually use logical cues?. There’s an entire body of educational science dedicated to this very topic….

How is this a thing?????

Posted by Sam ·
Status: Fixed2 comments

Comments2

  • Jason From Memo

    Team•

    Feb 18

    Hey,

    Thank you for bringing this to my attention, and I apologize for the inconvenience.

    I have just implemented a fix for this issue. Additionally, we have released Memo 8.0, which addresses many of these problems. Could you please refresh the page and check if it’s working for you now? I would appreciate your feedback on this.

    More information here:

    https://feedback.memo.cards/changelog/introducing-memo-80-or-biggest-release-ever

    Best regards,

    Jason

    Co-founder of Memo

  • Sam

    •

    Jul 24, 2025

    **Confounding responses