Stop Using AI as a Tool: Start Experimenting Like an Inventor
Most people treat AI like a calculator. They have a task. They ask the machine. The machine produces something. Done.
That’s practical. That’s useful. I use it that way too. But there’s a completely different game playing out right now—one that most people haven’t even noticed because they’re too busy checking boxes.
What happens when you stop using AI as a tool and start using it as an experiment?
That’s the question I’ve been exploring for the past year, and it’s fundamentally changed how I think about this technology.
The Calculator Approach Isn’t Enough Anymore
The calculator approach is simple: “Can you do this for me?” Write an email. Summarize a document. Generate some ideas. Get output. Move on.
There’s nothing wrong with that. It’s efficient. It’s exactly why AI has exploded so fast across every industry. People see the productivity bump and run with it.
But here’s what most people miss: The real power of AI isn’t in what it can do for you. It’s in what you discover when you start challenging it.
What Actually Happens When You Experiment
When you shift from using to exploring, your questions change. Instead of “Can you help?” you start asking “What happens if I try this?” and “Can you handle that?”
That’s where it gets interesting.
Over the past year, I’ve spent less time using AI and more time playing with it. Testing odd prompts. Pushing it outside the obvious use cases. Giving it constraints it wasn’t designed for.
Sometimes the results are genuinely impressive. Other times? Nonsense. And the nonsense is just as valuable—because the moment AI fails, you learn something real about how it actually thinks (or doesn’t).
Here’s What AI Is Ridiculously Good At
- Generating dozens of ideas when you ask for ten
- Rewriting content in five different styles, instantly
- Summarizing complex information into simple explanations
- Connecting patterns across unrelated topics
- Explaining concepts at any level of complexity
These things? AI nails them. Ask for five ways to approach a problem, and you’ll get thirty variations before you finish your coffee.
Here’s Where It Gets Weird
Now flip it. Give AI a task that requires strict rules, spatial awareness, or real-world constraints and something shifts. Not always. But often enough that it’s fascinating to watch.
The public debate about AI swings between two extremes: “It will solve everything” or “It will replace everyone.”
You don’t see the actual truth—the weird middle ground where AI is shockingly powerful and surprisingly fragile—unless you start experimenting with it yourself.
How to Actually Understand AI (Hint: Don’t Just Use It)
If you want to know what AI can really do, stop using it and start challenging it. Here are some experiments worth trying:
- Write a story in the voice of someone you admired as a kid
- Invent a game from scratch
- Explain something complicated to a curious twelve-year-old
- Turn a personal memory into a song
- Try to beat it at Scrabble
Some attempts will work beautifully. Some will fail spectacularly. Both teach you something crucial. And occasionally, you’ll stumble into something you didn’t expect at all—something that feels like a genuine collaboration rather than a command-and-response transaction.
Why This Matters Right Now
Most conversations about AI focus on productivity or job displacement. Those are important. But they miss what excites me most about this technology right now: curiosity.
AI rewards people willing to experiment. You try something. You see what happens. You refine and try again. The dynamic shifts from “here’s what I need from software” to “here’s what we can explore together.”
That’s when a tool becomes a playground.
That’s when you stop being a user and start being an inventor.
For 30+ years, I’ve watched technology reshape how we work and think. I’ve built businesses, launched products, and watched platforms rise and fall. What’s happening with AI right now feels different—not because of the technology itself, but because of how people are starting to interact with it. Those who experiment will move faster. Those who just use it will get left behind.
If you’re curious about pushing AI in unexpected directions—and seeing what actually happens when you do—that’s what I dive into on AI for Everyone, my weekly show. We don’t debate the future of AI there. We live it. We test it. We challenge it.
But here’s my bigger challenge to you: Stop thinking of AI as a tool. Start thinking of it as a partner in experimentation. Ask it something nobody else has asked. Push it outside the obvious. And when it surprises you—which it will—dig deeper.
That’s where the real future of AI happens. Not in the press releases. Not in the hype. But in the moment you ask a question nobody thought to ask and watch what emerges.
