better-ed
HomeUse CasesPricingPrivacy & TrustFAQContactAbout
LoginSign up for free
better-ed

Empowering educators with AI-powered voice assessment tools. Hear every student think—without doing more work.

LinkedInFacebookYouTube
CompanyAboutTeaching TipsBlogUse CasesPrivacy & TrustPricingFAQChangelog
Finland

Lapinlahdenkatu 16, 00180 Helsinki

Philippines

Unit 3006, One Corporate Centre, Julia Vargas Ave., Ortigas Center, Pasig City

© 2026 better-ed. All rights reserved.

Privacy PolicyTerms of ServiceCookie Policy
All tips
AI PROMPTS

Have it fade the worked examples

Full example, then examples with one missing step, then two, then a bare problem. AI builds the whole staircase from one solution.

Aug 25, 2026 · 1 min read

The jump from studying a worked example to solving alone is where novices fall. Fading bridges it: successive examples with more and more steps removed, so completion demands grow gradually; faded sequences outperform the abrupt example-then-problem switch (Renkl & Atkinson, 2003). Producing four versions of every example used to be too much prep. It's one prompt now. (Manning's worked-example-fading-designer skill formalises the same move.)

In practice

prompt.txt
Take this worked solution and produce a fading sequence:
Version A: fully worked.
Version B: final step blank.
Version C: last two steps blank.
Version D: problem only.
Keep numbers different in each version but difficulty identical.
  • Fade from the end backwards: students complete the last step first, which keeps them finishing rather than starting stuck.
  • Check the AI's arithmetic in every version before printing (see "Verify AI output like a student answer").

Quick takeaway: don't jump from example to problem. Generate the staircase between them.

More on ai prompts

AI PROMPTS

Ask for a task they can start, not finish

Productive failure needs a precisely calibrated task: solvable in part with what they know, unsolvable without what you are about to teach. Let AI calibrate it.


1 min read
AI PROMPTS

Make every wrong option confess a misconception

A multiple-choice question earns its keep when every wrong answer tells you exactly which misconception produced it, and which repair comes next.


1 min read
AI PROMPTS

Have AI plant why-questions inside the worked example

Students read worked examples passively. Ask the AI to plant a why-question at each tricky step and reading turns into explaining.


1 min read