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
CompanyAboutBlogUse CasesPrivacy & TrustPricingFAQChangelog
Philippines

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

Finland

Lapinlahdenkatu 16, 00180 Helsinki

© 2026 better-ed. All rights reserved.

Privacy PolicyTerms of ServiceCookie Policy
All tips
AI INTEGRITY

Assess the chat, not just the essay

On AI-open tasks, the transcript is where the thinking lives. Ask for it, and grade the questions they asked.

Sep 9, 2026 · 1 min read

If a task is AI-open, the interesting evidence isn't the polished product; it's how the student steered. A transcript shows whether they interrogated, refined, and rejected, or pasted the task and took the first output. Making the chat a submitted artefact turns prompting from a private shortcut into an assessable skill. (Manning's prompt-literacy-sequence-designer skill treats prompting as exactly this kind of teachable genre.)

In practice

  • Add one rubric line for the transcript: did the student push back, ask for alternatives, verify a claim, or change direction at least once?
  • Showcase two anonymised transcripts side by side: one that steered, one that accepted. The difference teaches itself.
  • Pairs with the disclosure norm (see "Ask how they used AI, not whether"): disclosure says what happened, the transcript shows it.

Quick takeaway: on open tasks, the conversation is the coursework.

More on ai integrity

AI INTEGRITY

Require one doubt, one check, one source

A standing rule for every AI answer: one place it could be wrong, one way to check it, one non-AI source to consult.


1 min read
AI INTEGRITY

Have students log what the AI did for them

After an AI-assisted task, one honest question: which parts could you now do alone? The answer calibrates them better than any rule.


1 min read
AI INTEGRITY

Make the AI argue against itself

Ask for the answer, then ask for the strongest case that the answer is wrong. The gap between the two is where the judgement lives.


1 min read