08Arc 3 / Extend & Keep / REPEAT / 120 MIN

Turn Good AI Use Into a Repeatable Work System

I can capture a successful AI method, test it on a new input, improve the reusable method, and leave with a personal playbook I can use again tomorrow.

Session / includes prompting practice Prompting exercise · use AI/GPT
What you'll learn

I can capture a successful AI method, test it on a new input, improve the reusable method, and leave with a personal playbook I can use again tomorrow.

Bring / required

Required: computer or phone with AI/GPT for the live prompting route. Bring your own small work task or use the starter case. Your course playbook should be open for notes.

What you'll do

Follow the session exercise, run only the text explicitly marked as a prompting activity, inspect the result and record evidence in your session card.

Take away

A completed session card, a checked result or decision, and an update to your Personal AI Work Playbook.

Explanations and workbook templates are not prompts. Prompting activities are labelled explicitly.

The session

THREE BITES. ONE PRACTICAL MOVE.

08ACapture What Worked
08BRun It Again: Fixed vs Variable
08CBuild My Personal AI Work Playbook

What you'll practise

  • Capture the method behind a successful result.
  • Test the method again on a new input.
  • Turn the strongest methods into a personal AI Work Playbook.
Keep this idea

A repeatable workflow is more valuable than a lucky prompt.

Try it with the supplied case

Prompting exerciseYou write or revise the AI instruction as part of this exercise. Only the instruction for that step plus the referenced case material goes into AI; the worksheet remains your notes.

Question: Can I use this again without rebuilding the conversation? Input: your accumulated playbook, Q-105 or a new own-work input.

  1. In 12 minutes reduce the successful method to inputs, steps, checks and human boundary.
  2. In 12 minutes give it to a partner who follows it without extra explanation. Use Q-105 if they have no input; mark where clarification is needed.
  3. In 12 minutes revise the method and assemble existing evidence. Do not introduce a new stack of frameworks in the capstone.
  4. Use the closing 10 minutes to compare baseline/final evidence, save/export and answer the opening question with an unresolved gap and next application.

Keep/check: A partner can identify current source, calculate RM750, preserve unconfirmed pickup and repeat the useful method. Keep output plus evidence, not confidence alone. Optional transfer: try one new input next week.

Use this now

SESSION 8 PARTICIPANT CARD

Use this as your session notes and evidence. Do not paste the whole card into AI. Only text explicitly labelled Prompting exercise belongs in an AI chat.

AI WORK PATTERN

WHEN:
______________________________________________________

INPUT:
______________________________________________________

AI JOB:
______________________________________________________

OUTPUT:
______________________________________________________

CHECK:
______________________________________________________

HUMAN:
______________________________________________________

SECOND-RUN REVISION RECORD

What drifted?
______________________________________________________

What failed?
______________________________________________________

What instruction / check was added?
______________________________________________________

What remains human?
______________________________________________________

Playbook section to keep this under:
______________________________________________________

PROOF
Could another person repeat the method without reading your chat?
YES / NO

What would they still need?
______________________________________________________

RETURN TO MY QUESTION
Status: answered / demonstrated / needs more evidence / outside this course
Evidence or gap:
Next application:

Class rhythm

Feel itExperience the problem or contrast before naming the method.
Name itMake the useful pattern explicit.
Use itApply it to a real work object and inspect the result.
Keep itCapture the part worth reusing after class.
Final course evidence

FINAL PROOF RECORD

Use this to show the complete before → after → second run → human boundary journey. Keep the checked output and explain what the second input taught you.

FINAL PROOF RECORD · WORK WITH AI

ANCHOR TASK
______________________________________________________

BEFORE

Normal method used:
______________________________________________________

Original output weakness:
______________________________________________________

Score:
Usefulness ___/5   Ready ___/5   Confidence ___/5

AFTER

Improved method used:
______________________________________________________

Evidence of improvement:
______________________________________________________

Score:
Usefulness ___/5   Ready ___/5   Confidence ___/5

SECOND RUN

New input used:
______________________________________________________

Did the method survive?  YES / NO

What drifted?
______________________________________________________

What stayed fixed?
______________________________________________________

HUMAN BOUNDARY

AI may:
______________________________________________________

Human must:
______________________________________________________

Stop if:
______________________________________________________

NEXT MOVE

The next AI capability I want to strengthen is:
______________________________________________________
← Previous / 07Create & Work Multimodally Programmes →Explore the three courses