If ChatGPT Isn't Working, the Problem Might Be You

This morning, I said hi to a friend. She replied with this "prompt":

—"I’m going to have breakfast and then heading to the café."

And I thought: “Nice, she’s working from a coffee shop today. Boho vibes, laptop, and a flat white.”

Five minutes later, she hits me with this "prompt":

—“Did you send the email?”

And I’m just sitting there like… What email?

My brain started mentally scrolling, flipping through every open tab in my mind, and then—BOOM! There it was: three days ago we had talked about emailing a specific café where we’re planning to host an event.

And in that moment, I realized something as obvious as it is painful:

It wasn’t a café. It was that café. The meeting café. The email café.

The context-she-didn’t-give-me-but-somehow-expected-me-to-magically-guess.

This is pretty much how most people write prompts in ChatGPT.

We keep talking to people as if they live inside our heads. And they don’t.

Not your boss. Not your partner. Not the guy at the corner store. And ChatGPT? Even less.

We speak (and write prompts) in fragments. We leave half the message stuck in our neurons and expect the other person to complete it, guess it, translate it with full accuracy—and be grateful about it, too.

Spoiler 2: That’s not how communication works. It never has.

We’ve heard it a thousand times: communication is key.

In work. In relationships. In life.

And yet… we miss the mark more than a drunk guy at a county fair shooting range.

You sit in front of ChatGPT, type a vague prompt, and then get disappointed with the result.:

“This isn’t helpful.”
“That’s not what I meant.”
“This is too generic.”
“That’s not even close.”

And of course, the problem is the AI. Right?

Are you sure?

Because maybe you didn’t give it any context.
Maybe you didn’t tell it you’re an instructional designer creating a course for a specific audience.

Maybe you forgot to mention you want a tone that’s friendly yet professional, that the final output goes to a corporate client, that it needs to be creative, full of examples, and have an ending that makes your reader want to send you flowers.

Maybe you didn’t say any of that.

Maybe you just typed:
"Write me an e-learning module on leadership."
And expected magic.

ChatGPT showed up to point out where your communication sucks.

Yes, I said it.

And if that stings a little, good. It means something’s starting to shift.

GPT is the pebble in the shoe of a generation that got used to vague briefs,
emails that say “we discussed this in the meeting,”
and instructions like “make it look like Canva but more pro.”

In instructional design (and marketing, and life), one thing can’t be skipped: analysis.

The kind that sniffs out the context.
That understands the user.
That sharpens the questions.

If you don’t know what you need, you won’t know how to ask for it.
And if you don’t ask properly, don’t blame the messenger (or the algorithm).

So, if ChatGPT isn’t giving you what you want, before saying “this tool sucks,” ask yourself:

• Am I giving enough context?

• Am I being clear about my objectives?

• Do I actually know what I want?

• Am I expecting it to think for me?

Because sometimes the problem isn’t that the tool isn’t smart enough.

Sometimes—just sometimes—the problem is you.

That part of you that still hasn’t learned how to communicate clearly.
That inner voice that assumes “the other person will just get it.”
That habit of blaming external things because looking inward… is uncomfortable.

ChatGPT doesn’t take your job. It is your copilot.

It takes away your excuses.

And if you’re resisting it, maybe it’s not because it doesn’t work—
Maybe it’s because it’s holding up a mirror to the very thing you’ve been avoiding.

Breathe. Put in a little effort.
Learn how to ask properly.

And then, yes...
Let the magic happen.