Stop Asking AI Questions - Give It Problems to Solve With You
Over the past two weeks, I've shared how to get more thoughtful responses from AI by demanding deeper thinking, and how to recognize when a conversation isn't working so you can start fresh.
Today I want to talk about something that comes even before those strategies: how you set up the conversation from the very beginning.
Here's what I notice: most people use AI like a search engine. They ask a question, get an answer, move on.
But AI isn't a search engine. And when you treat it like one, you get search engine results - generic, surface-level, often missing what you actually need.
Questions vs. Problems
When you ask AI a question, it gives you an answer. When you give AI a problem to solve with you, it gives you thinking.
The difference matters from the very first exchange.
Let me show you what I mean.
Question approach: "What are effective teaching strategies?"
AI gives you a list. Five strategies. All generic. Nothing you couldn't find in any education textbook. You get information, but not insight.
Problem-solving approach: "I'm teaching a graduate course on information systems. My students come from diverse backgrounds - some technical, some not. They're struggling to connect theory to practice. Help me think through how to bridge that gap."
Now AI has context. It can tailor responses to your actual situation. It asks clarifying questions. It suggests approaches specific to your constraints. You get a conversation, not a list.
Why This Changes Everything
Remember when I talked about demanding thinking from AI? That works much better when you've set up the conversation as collaborative problem-solving from the start.
Remember when I talked about recognizing when conversations go off track? When you begin with clear problem framing, conversations are far less likely to derail in the first place.
This is the foundation that makes everything else work better.
What AI Needs to Help You
When you frame something as a problem to solve together, you're giving AI:
Context about your specific situation
Constraints that actually matter
What you've already tried or considered
What success looks like for you
AI can't read your mind. When you just ask questions, it has to guess at context. When you share the actual problem, it has something real to work with.
The Basic Structure
Instead of asking "How do I [X]?" try this:
"Here's my situation: [context] Here's what I'm trying to accomplish: [goal]
Here's what makes this challenging: [constraints] Help me think through possible approaches."
Real Examples
Before: "How do I give difficult feedback?"
After: "I need to give feedback to a colleague who's missing deadlines. They're talented but seem overwhelmed. I want to address the issue without damaging our working relationship or making them defensive. Help me think through how to approach this conversation."
Before: "What's a good exercise routine?"
After: "I'm 46, haven't exercised regularly in years, dealing with some knee issues. I want to build strength and energy but I'm worried about injury. I have 30 minutes most mornings. Help me think through what might be realistic and sustainable."
The second version gives AI something to actually work with.
What You'll Notice
When you shift from questions to collaborative problem-solving:
Responses become specific to your situation rather than generic AI asks better follow-up questions to understand what you actually need
You get approaches tailored to your actual constraints The interaction feels less like interrogation and more like thinking together
And when you do need to push back and demand deeper thinking (week 1), or when you need to abandon and restart (week 2), you're doing so from a much stronger foundation.
Start Here
Next time you're about to ask AI a question, pause. Reframe it as a problem. Give context. Share constraints. Invite collaboration.
Set up the conversation right from the beginning, and everything that follows works better.
