How I Stopped Treating AI Like a Search Engine
At the start of 2024, I was still in the habit of treating AI like a search engine - asking a question and hoping to get the “right” answer. The trouble with that approach is it often comes with a closed mindset; we typically already have an answer in mind.
Through trial and error, I’ve learned that getting value from AI enabled tools requires curiosity and an open mind. Adopting an “AI mindset” means intentionally practicing better questioning and allowing the technology to serve as a true thought partner.
Growth Mindset: A Key Ingredient
Microsoft CEO Satya Nadella famously credits Microsoft’s success in AI to adopting a Growth Mindset: a shift from chasing competitors to setting the pace. He encourages everyone at Microsoft to be learn-it-alls, not know-it-alls.
Inspired by this, I realized my own mindset was affecting the quality of outputs I received from AI tools like ChatGPT and Copilot. Once I started asking more open-ended questions and giving clearer context, I saw a significant improvement in the responses I got.
My Approach: Purpose, Context, Conversation
Define Your Goal
Before I chat with AI, I clarify what I’m trying to achieve. What's the task I need AI to help me with? What am I trying to accomplish? What does success look like? I also gather what I already know or suspect about the topic, identify relevant sources or examples, and note any helpful references.
Treat it like a Conversation
Say you walked up to a friend and ask them to help you, but their response surprised you (not in a good way). You wouldn’t just turn around and walk away. You’d clarify what you meant. There’s a reason you went to them for help, right?
I went from rolling my eyes at the outputs I’d get when I first started using ChatGPT to now turning to it multiple times a day—simply by improving the context I shared.
Provide Persistent Context
Customize Your AI: In ChatGPT, for example, you can set persistent context in the “Customize ChatGPT” settings.
Contextual Prompts: In each interaction, I specify the role the AI is playing, the project I’m working on, and any reference documents or URLs it should consider.
Encourage Clarifying Questions
One request I regularly use that I’d never ask a friend is:
“Before you answer, please ask me clarifying questions to ensure you have all the context you need. Then, rewrite my prompt for a better output.”
This encourages the AI to engage proactively and helps me refine my prompts.
Evaluating Outputs
When a project relies on data or facts, I always ask for sources. Then I evaluate whether those sources are trustworthy. If I need high accuracy, I’ll triangulate the information by checking multiple reputable references.
Does more than one trustworthy source say the same thing?
I also reflect on what I already know and consider what might be missing or biased. Asking myself “Who benefits from me thinking this way?” helps me catch blind spots. Sometimes, I even enlist another AI tool to review outputs for bias or oversights.
Shifting My Mindset
This active, iterative approach has changed my perspective. I went from feeling uncertain about the usefulness of generative AI to beginning new projects by asking, “How can AI help me do this better?”
Now, at the start of 2025, here’s what I teach others who want to become AI-enabled at work:
We can’t approach generative AI like a search engine.
Meaningful outputs require a learn-it-all mindset, and three key skills are vital:
Conversation – Treat each interaction like a dialogue (not one and done).
Critical Thinking – Validate sources and check biases.
Storytelling – Use context and examples to convey ideas clearly.
Learning Resources
I’d love to explore more tools and insights. Can you recommend any books, courses, podcasts, or other resources that might help me (and others) continue to grow in these three areas—conversation, critical thinking, and storytelling?
