Treat the artifact as a starting point

A polished demonstration shows that a system produced an output. By itself, it does not reveal which parts the learner understands or who supplied them. That is not a reason to dismiss the work. It is a reason to ask a few specific, friendly questions and keep contributions visible. Invite the learner to choose a part they want to discuss. Ask what it is for, what goes in and what comes out. Avoid beginning with an accusation about AI use or asking for a performance designed to catch them out.

Try explanation and a small change

Choose one design decision, such as how images are grouped or how an input is prepared. Ask why that choice was made and what another option might change. Then propose one manageable requirement: a darker photograph, a new category or an empty input. Ask the learner what they expect before running the example. Compare the expectation with the result and discuss what remains uncertain. The aim is to make reasoning visible, not to demand perfect terminology. A learner can demonstrate an emerging understanding while still needing prompts or help.

A learner explaining a code decision on a whiteboard
MAKE UNDERSTANDING VISIBLE

The finished output is the beginning of the conversation.

  1. 01Explain a decision
  2. 02Change one requirement
  3. 03Investigate the result

Rebuild a bounded component

Ask whether the learner would like to reconstruct a small function, diagram or test from a simpler starting point. Keep the scope narrow and agree what references or assistance are allowed. Record the support used: a mentor hint, documentation, starter code or an AI suggestion. Do not infer mastery from a single success or inability from a single difficult attempt. Fatigue, unfamiliar instructions and the social setting can affect a conversation. Repeated ordinary observations within lessons are more useful for planning than a one-off household examination.

Make attribution part of progress

A short contribution strip can name what the learner attempted, what the mentor scaffolded and where AI assisted. Use descriptions rather than ownership percentages. “Explained categories and tested unfamiliar images” is more informative than a claim that work is entirely independent. If attribution is unknown, preserve that uncertainty and ask how it can be recorded next time. Learning Prospect’s evidence kit creates prompts and a notes sheet; it does not detect AI authorship, certify competence or assess a child automatically. Use the conversation to choose the next useful teaching step and celebrate specific effort without inventing a broader outcome.

PUT THIS INTO PRACTICE

Take a clearer brief to the next conversation.

Open the related tool

Sources and limitations

Editorial decision guidance, not an assessment of an individual learner or provider. No lessons were observed. Linked sources provide context, not endorsement of Learning Prospect.

Source review: 18 November 2026. Provider terms can change. Read our methodology.

Continue reading

Plan a six-week AI project for a teenagerShould you renew or change your child’s AI tutor?