ChatGPT made you an expert? Cute, but nope
Let me start by saying this: I’m not anti-ChatGPT.
It can simplify processes, spark ideas, and save you hours. I use AI models too.
But here’s the thing — people are starting to treat it like a shortcut to expertise. And that’s where things get murky AF.
What ChatGPT is (and what it isn’t).
ChatGPT (and its pals like Claude, Gemini, Copilot, Perplexity, etc.) isn’t a search engine. It doesn’t crawl the internet in real time like Google.
Instead, it’s a large language model (LLM) — which is just a fancy way of saying it predicts the next word in a sentence based on patterns it’s learned. What it “knows” comes from:
its original training data (big dumps of text from books, articles, websites, forums)
whatever “high-authority content” it’s allowed to draw from now (think directories, “best of” roundups, visible blogs, and podcasts)
and the people who know how to play the content game
So when you ask it something, it doesn’t know the answer. It guesses what a convincing answer looks like, based on those sources.
Sometimes that’s helpful. Sometimes it’s neutral. And sometimes it’s biased, misleading, or flat-out wrong.
But because ChatGPT delivers everything with confidence, most people don’t stop to question it. They just scan the response, think: ‘shit I sound good’, copy, paste, do a little edit, remove the em dashes (more on that another day), and post.
Why we overestimate what ChatGPT (and AI models in general) can do:
Enter the Gell-Mann Amnesia Effect. It’s this annoying little cognitive bias phenomenon.
Coined by Michael Crichton (yes, weirdly, the Jurassic Park guy) to describe how we blindly trust media we should know better than to trust. In short:
You read something in your area of expertise, spot all the mistakes, and roll your eyes. Then you flip the page… and trust everything else in the paper as if it’s gospel.
Today, we do the same thing with digital media and AI.
We spot the flaws in our lane - but assume it must be brilliant everywhere else.
That’s why you’ll hear things like:
Website designers say, “ChatGPT’s hopeless for layouts that support the customer journey, but incredible for writing website copy.”
Copywriters say, “AI’s destroying thought-leaders and brand voice, but amazing at dishing out Squarespace code.”
Coaches say, “I’d never use it for mindset work, but hey, it’s got great nutrition advice.”
See the pattern? We dismiss it where we know better, then give it a free pass where we don’t.
And that’s risky. Because the better AI gets at sounding right, the harder it is to spot when it’s wrong.
When we use ChatGPT outside our lane, we don’t just risk putting dodgy info into the world…
We water down credible human expertise.
Take me, a copywriter, teaching Squarespace development tips because ChatGPT told me how to save a client a few dollars? Hard no.
If I wanted to share that, I’d collaborate with a trusted Squarespace designer / developer who actually knows their shit.
And the reverse is true. A website designer dishing out “copywriting prompts” without understanding buyer psychology? Risky.
A life coach who’s suddenly supporting a whole heap of women in their 50’s starts dishing out menopause nutrition advice thanks to an AI summary? Dangerous.
So, what do we do?
Use ChatGPT (and other AI models) as a tool, not an authority.
If you’re going to use it, use it responsibly:
Keep it in your lane. Let it tidy or repurpose your own expertise, not replace it.
Critically analyse what it spits out - don’t just accept it as fact because it sounds right.
When you want to share expertise outside your zone of genuis? Collaborate. Bring in the humans who live and breathe that work.
Because the more we flood the internet with ChatGPT’s persuasive tone as proof of accuracy, the harder it becomes to find the voices that actually know their shit — and the easier it is for biased, misleading, or harmful info to spread unchecked.
Yes, ChatGPT can be helpful.
Use it to simplify, streamline, or get unstuck.
But your experience, your judgement, your ethics — that’s the bit ChatGPT can’t replace.
So by all means, use it as a tool.
But don’t mistake it for expertise.
A quick “am I using ChatGPT responsibly?” checklist:
Lane check: Am I qualified to give this advice? If not, who is?
Source check: Where did this fact come from? Can I back it up?
Regency check: Is this up-to-date for my region/industry, or could it be pulling from outdated info?
Bias check: Is this re-saying what the loudest sites say, or adding something real?
Impact check: Could someone be misled or harmed if I get this wrong?
Hey cutie - I’m Pip, the creative and strategic website copywriter over here at Articulate Communications.
I’ve been told once or twice: I’ve got a way with words. And I sure as sugar know how to help you find your voice, share your stories, connect with your people, and sell while you sleep (or sip 2pm chilli coconut margs in the sunshine).
It’s my job to help you finally move from *meh* to memorable brand. And I do that with a saucy little blend of creative and strategic copywriting.
Wanna see what that looks like for you? Holla at me.