BindlCorp · Beginner’s Guide to AI · Vol. 2
March 2026 · bindlcorp.com · 20 min read + activities
In Vol. 1, you opened the tools. You had a real conversation. You probably got at least one response that surprised you — and maybe one that missed the mark entirely. Both of those things are useful information, and this volume is about understanding why they happened.
The difference between a mediocre AI response and a genuinely useful one almost always comes down to how you asked. Not the tool. Not the model. How you framed the question. This is a learnable skill — and once you have it, every AI interaction gets substantially better.
Three activities this volume. By the end, you’ll know how to get consistently useful answers, how to use AI for research without getting burned by bad information, and how to build habits that make this stick. Let’s go.
Just joining us?
Vol. 2 builds on the tools and concepts from Vol. 1 — Claude, ChatGPT, and how to treat AI like a conversation rather than a search bar. If you haven’t done that one yet, it’s worth 20 minutes first.
← Start with Vol. 110x
Better results from the same tool with better prompts
3
Techniques that cover 90% of what you’ll ever need
0
Technical knowledge required to do any of this
Why this matters
The One Thing That Changes Everything
Think about the last time you asked someone for help and got a vague or unhelpful answer. Usually it wasn’t that the person was incapable — it was that they didn’t have enough information to help you well. AI works the same way. It can only work with what you give it.
Most people type short, bare questions: “how do I write a cover letter” or “explain investing.” They get generic responses. Then they think the tool isn’t that useful. The tool isn’t the problem.
There are three things you can add to almost any question that will dramatically improve the answer. You don’t need to use all three every time. But understanding them changes how you interact with these tools permanently.
01 · Role
Tell it who you are and who it should be. “I’m a nurse explaining this to a patient” produces something completely different than the same question with no context.
“Act as a plain-English financial advisor…”
02 · Goal
Say what you’re actually trying to accomplish — not just what you’re asking for. “I want my landlord to take me seriously” is a goal. “write me an email” is a task.
“…I need to sound firm but not aggressive”
03 · Format
Tell it how you want the answer. A bullet list, a short paragraph, a step-by-step walkthrough, a table. If you don’t say, it guesses — and the guess isn’t always right.
“Give me three short bullet points.”
See the difference
What most people type
“how do I ask for a raise”
Gets: a generic five-step article you could have Googled. No context. No specificity. Not actually useful.
What actually works
“I’m a project manager at a mid-size company, been in this role for 2 years, just led a product launch that went well. I want to ask my manager for a raise at my review next week. She’s direct and data-driven. Help me make the case in a way that doesn’t feel awkward. Give me the key points I should make, and what to say if she pushes back.”
Gets: a response built for your actual situation. Specific talking points. Real pushback prep. Something you can use tomorrow.
Hands-on session
Three Activities. Real Results.
Activity 1 of 3 · ~7 minutes · Claude or ChatGPT
The Upgrade — Take a Bad Prompt and Fix It
Open Claude or ChatGPT. You’re going to run the same question twice — once the way most people do it, and once the right way. The difference will be obvious.
Step 1 — Type this exactly:
“how do I get better at public speaking”
Read the response. It’s probably fine. Probably generic. A list you’ve seen before. Now do it again — same window, new conversation — with this version:
Step 2 — Now type this:
“I’m a [your job] who has to give a presentation to [your audience — coworkers / clients / my team] in [timeframe]. I’m not afraid of presenting but I tend to rush when I’m nervous and lose my train of thought. I don’t want generic tips — I want 3 specific things I can practice this week that will actually help with those two problems specifically.”
Fill in the brackets with your actual situation. Then read both responses side by side.
What to notice
The second response knows your job, your audience, your timeline, and your specific problem. It didn’t need more AI capability to get there. It needed more information. That’s the whole lesson — and you can apply it to anything.
Activity 2 of 3 · ~8 minutes · Claude recommended
The Research Test — Use AI Without Getting Burned
AI is useful for research — but it can be wrong with complete confidence. It doesn’t hedge the way a cautious person would. It doesn’t say “I think” the way a friend might. It just answers. This activity teaches you the two-step that keeps you safe.
Step 1 — Pick a real question you actually have. Something like:
“What’s the difference between a Roth IRA and a traditional IRA and which makes more sense for someone in my situation?”
“I’m having [symptom]. What are the most likely causes and what should I actually pay attention to?”
“What are my rights if my landlord hasn’t fixed [issue] for [timeframe] in [your state]?”
“What should I know before [buying my first car / negotiating a salary / starting a small business]?”
Step 2 — After you get the answer, ask this as a follow-up:
“What parts of what you just told me are most likely to be wrong, outdated, or vary depending on my specific situation? What should I verify before acting on any of this?”
Read what it says. This is AI being honest about its own limitations — and it’s genuinely useful. It’ll often flag exactly where you should double-check with a professional or a current source.
The rule going forward
Use AI to get oriented fast and ask better questions — then verify anything that matters before you act on it. That combination makes you more informed than most people walking into a doctor’s office, a financial advisor’s office, or a negotiation. You’re not replacing expert judgment. You’re going in prepared.
Activity 3 of 3 · ~5 minutes · Any tool
The Follow-Up — Where the Real Value Lives
Most people ask a question, read the response, close the tab. That’s the weakest way to use AI. The follow-up is where it gets interesting — when you push back, go deeper, or redirect. This activity is short because the point is simple. It just needs to become a habit.
Go back to the response from Activity 1. Then try at least two of these:
“That second point — can you go deeper on that specifically?”
Forces it to expand on the part that was actually useful to you instead of the whole response.
“Assume I have no time this week. What’s the single most important thing from everything you just said?”
Makes it prioritize for you. Useful when a response is comprehensive but overwhelming.
“I disagree with the part about [X]. Here’s my thinking. What am I missing?”
AI will push back thoughtfully. It’s not just agreeing with everything you say. This is where it starts feeling like a real conversation.
“Now rewrite your answer as if I’m explaining it to someone with no background in this at all.”
Instantly changes the register. Useful if the response was too technical — or if you need to explain something to someone else.
Making it stick
How to Build the Habit Without Thinking About It
The people who get consistently useful results from AI aren’t running through a checklist every time they open the tool. They’ve built a reflex. They give context automatically. They follow up by default. They know when to trust the output and when to verify it.
That reflex comes from repetition, and it forms faster than you’d think. Here’s the simplest version of the habit loop that works:
01
One real thing per day
Not a test. Something on your actual to-do list. An email, a decision, a question you’ve been putting off. One thing. Every day for a week.
02
Always add context
Before you hit send on any prompt, ask yourself: does this have my role, my goal, and the format I want? Takes five seconds. Changes the response significantly.
03
Always follow up once
After the first response, send one follow-up — push deeper, ask it to simplify, or tell it what missed. Never close the tab after the first answer.
Seven days of that and it stops being a technique. It becomes how you use the tool.
Keep this handy
The Prompt Cheat Sheet
These are the structures that cover 90% of what most people need. You don’t memorize them — you steal them and fill in the blanks.
For getting advice
“I’m a [who you are] dealing with [situation]. My goal is [what you want to happen]. What should I do, and what am I probably not thinking about?”
For writing help
“Help me write [type of message] to [who]. The context is [situation]. I want to come across as [tone — professional / firm / warm / direct]. Give me a draft and then three alternatives with different approaches.”
For understanding something complex
“Explain [topic] to me. I know [what you already understand]. I don’t understand [specific part that’s confusing]. Use an example from [something I’m familiar with — my industry / everyday life / a specific situation].”
For decisions
“I’m deciding between [option A] and [option B]. Here’s my situation: [context]. What are the strongest arguments for each? What would you ask me to figure out before deciding? What are most people in this situation wrong about?”
For research
“Give me an overview of [topic] — what I need to understand, what’s contested, and what questions I should be asking before I [make a decision / talk to a professional / take action]. Then tell me what parts of your answer I should verify independently.”
Before you close this tab
One Thing to Do Today
Pick one thing from your prompt cheat sheet above. Pick an actual situation you’re in right now — something real, not a test. Fill in the blanks with your actual details and send it.
Then follow up once. That’s it. You now have the core of the skill. Everything else is just more repetitions.
Recommended for this volume
claude.ai →
Best for nuanced conversations and research
Also solid
chat.openai.com →
Direct, structured, good for drafting
Coming in Vol. 3
Using AI at Work Without Making It Weird
How to integrate AI into your actual job — meetings, emails, reports, decisions — without it being obvious, without it replacing your judgment, and without your boss thinking you’ve stopped thinking. Practical, specific, and aimed at people who already have jobs they want to do better.
Free Weekly
The AI
Digest
What actually happened in AI this week. What it means for your job, your money, and your life. Plain English. Every Friday. Free.
No spam. Unsubscribe anytime. Plain English. Every Friday.
Leave a Reply