Brief & Steer AI for Better Results
I can give AI clearer direction, improve weak results deliberately, and choose how I want AI to collaborate with me.
I can give AI clearer direction, improve weak results deliberately, and choose how I want AI to collaborate with me.
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.
Follow the session exercise, run only the text explicitly marked as a prompting activity, inspect the result and record evidence in your session card.
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.
THREE BITES. ONE PRACTICAL MOVE.
What you'll practise
- Turn an intention into a useful brief.
- Steer output through constraints, examples and feedback.
- Choose a useful interaction mode instead of always asking for an answer.
Better output usually begins with better direction — then deliberate steering.
Try it with the supplied case
Question: How do I improve a weak reply? Input: S1/S2 and Reply A.
- In 4 minutes mark the two most important defects in Reply A. Do not restart the entire task inventory.
- In 12 minutes write a brief naming receiver, purpose, facts, boundaries and output. Produce or rewrite a corrected draft.
- In 8 minutes make one targeted follow-up: remove an unsupported promise, improve the ask or adjust tone for the receiver. Compare before/after.
- Trainer demonstrates answer/draft, interviewer, critic and step-by-step collaborator modes; learners practise one unfamiliar mode for 8 minutes within Bite C.
Keep/check: Brief and one effective follow-up. Did the revision meet the actual objective? Tone may be the needed improvement; do not declare tone-only changes shallow. An unclear collaboration role is one possible cause, not the diagnosis of every weak output.
FIVE-PART BRIEF
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.
ANCHOR TASK ______________________________________________________ NORMAL INSTRUCTION I USED ______________________________________________________ FIVE-PART BRIEF Outcome: ______________________________________________________ Situation: ______________________________________________________ Material: ______________________________________________________ Boundary: ______________________________________________________ Deliverable: ______________________________________________________ STEERING MOVE USED [ ] Question it [ ] Challenge it [ ] Improve it What changed? ______________________________________________________ INTERACTION MODE [ ] Answer me [ ] Think with me [ ] Challenge me [ ] Interview me Why this mode? ______________________________________________________ What still needs human judgement? ______________________________________________________ PROOF One ambiguity removed: ______________________________________________________ One better result: ______________________________________________________ RETURN TO MY QUESTION Status: answered / demonstrated / needs more evidence / outside this course Evidence or gap: Next application:
Class rhythm
Field-guide bridge
Session 02 teaches the practical habit. The Prompt Engineering field guide is where I keep testing why certain instructions work, where prompt improvement stops helping, and how to compare changes without turning them into folklore.