Guides that turn AI curiosity into real-world skill

You are a [role/expert] with expertise in [domain].

Your task: [specific task description]

Context: [audience, constraints, format requirements]

Examples (if needed): [show 1-2 examples]

After completing the task, critique your own response and identify one way to improve it.

[Your question or problem]

Let's think through this step by step:
1. First, [step 1]
2. Then, [step 2]
3. Finally, [step 3]

Based on this reasoning, [conclusion]

[Initial request]

[After receiving response:]
"That's a good start. Now refine it by: [specific improvement 1], [specific improvement 2], and [specific improvement 3]."

[Continue iterating as needed]

I need ideas for [problem/challenge].

Please provide ideas from three different perspectives:
1. As a [perspective 1, e.g., "creative designer"]
2. As a [perspective 2, e.g., "practical engineer"]
3. As a [perspective 3, e.g., "business strategist"]

For each perspective, give one concrete, actionable idea.

Summarize the following [article/document/meeting notes] in [format]:

[Paste content]

Requirements:
- Length: [word count or bullet points]
- Focus on: [key themes or topics]
- Audience: [who will read this]
- Tone: [professional/casual/academic]

I've written the following [type of content]. Help me improve it while maintaining my voice:

[Paste your draft]

Please:
1. Identify 3 areas that need improvement
2. Suggest specific revisions for each
3. Keep my casual/professional/[your] tone
4. Maintain the original message and intent

I want to understand [concept/topic].

Please:
1. Explain it in simple terms (like I'm [age/background])
2. Provide a real-world analogy
3. Give me 3 practical examples
4. Suggest one hands-on exercise I can do to reinforce this

If I'm confused, I'll ask follow-up questions.

Does this task involve confidential, personal, or proprietary information that shouldn't be shared with an AI tool?

What are the consequences if the AI is wrong? Is this a high-stakes decision (hiring, medical, legal, financial)?

Could this AI application disadvantage certain groups? Does it involve human judgment about people or opportunities?

How will I verify the AI's output? Do I have the expertise to catch errors? Is there a human review process?

Should others know AI was involved? Am I required to disclose AI use (academic, professional, legal contexts)?

Does my organization have an AI use policy? Have I checked what's allowed and what requires approval?

What happens to the data I input? Will it be used for training? Can I delete it? Am I using the right version (e.g., Teams vs. consumer)?

If the AI fails or provides poor results, what's my backup plan? Am I overly dependent on this tool?

What specific problem am I solving or outcome am I creating? Be concrete and measurable.

How will I know this project succeeded? What does 'done' look like?

Which AI tools will I use? (ChatGPT, Claude, Midjourney, etc.) Are they free or paid?

How much time can I realistically dedicate? Break it into sessions (e.g., 3 sessions of 30 minutes).

List 3-5 concrete steps to finish this project. Make each step actionable.

What could go wrong? How will I handle setbacks or unexpected results?

Besides the final output, what do I want to learn from this project?

Who will I share this with? How will I document what I learned?