ChatGPT on Steroids

In February 2023, weeks after ChatGPT launched, I wrote a short post about using AI as a starting point — not a replacement. Here's what I got right, what I missed, and what I'd tell that version of myself now.

Three years ago, I published a blog post called "ChatGPT on Steroids for Your Social Media." It was February 2023. ChatGPT had been public for about two months, and the people who were paying attention were absolutely losing it. Teenagers were quietly using it to write their homework. Their parents had no idea it existed. The tech world was split between euphoria and existential dread, and most of the general public hadn't even heard the letters G-P-T strung together yet.

I was in a coffee shop when I first really sat with it. Mouth literally hanging open, staring at my screen, equal parts mind-blown and terrified. I knew right then that everything was about to change — I just didn't know how fast.

I found that post recently, still sitting on my WordPress site, completely untouched. Reading it back now?

It's like finding a diary entry from the first week of a relationship that ended up changing your life.

Screenshot of the original ChatGPT on Steroids blog post as it appeared on the Fran Jeanes WordPress site in February 2023, with the original styling and text.
The original post, exactly as it appeared in February 2023

The gist was simple. I was using ChatGPT to generate starting points for content — not finished products. I actually showed readers a paragraph I'd written myself, then the same paragraph rewritten by ChatGPT, and pointed out all the "marketing-speak" it injected. Words like cutting-edge and indispensable tool in your arsenal.

(Because nothing says "a human wrote this" like the word arsenal.)

My advice back then was straightforward: use it selectively, always go behind it and check, and rewrite enough that the output actually sounds like you wrote it. I also linked to a tool I'd built — a social media quote generator that paired inspirational quotes with stock photo search terms and ready-made captions.

I was proud of that little tool. It solved a real problem for the small business owners I worked with — the ones who knew they needed to post consistently but did not have two hours a day to sit around crafting content from scratch.

So what did February 2023 Fran get right?

The core instinct. AI is a starting point — it is not the finished product. That hasn't changed. If anything, it's become more true as the tools have gotten more capable. The people who let AI write everything for them?

They sound like AI.

The people who use AI to accelerate their own thinking sound like sharper versions of themselves. Big difference.

I was also right that you need to go behind it and check. In 2023 I was talking about factual errors and generic phrasing. In 2026, the checking is more nuanced — it's about voice, about specificity, about whether the output actually reflects what your business does differently from everyone else's. But the principle?

Identical.

The people who let AI write everything for them sound like AI. The people who use it to accelerate their own thinking sound like sharper versions of themselves.

Now here's what I missed.

What I didn't see coming — and honestly, I don't think anyone did in February 2023 — was how fast the conversation would shift from "use this chatbot to write captions" to "redesign your entire workflow around AI." I was thinking about content generation. I should have been thinking about process transformation.

That social media quote generator I built? Clever, sure. But it was a point solution. It helped with one task on one platform for one type of content. Now — sitting down with an owner-operator, mapping their entire business workflow, and finding the places where AI can remove friction, recover revenue, or protect against risk — that's a fundamentally different conversation.

The tool I built in 2023 was aspirin.

What businesses actually need is a diagnosis.

I also underestimated how quickly "sounds like a human wrote it" would stop being a differentiator and start being the bare minimum. In early 2023, just knowing how to prompt ChatGPT was a skill. Now everybody and their grandmother can prompt. The skill in 2026 is knowing what to build with it — and more importantly, knowing what not to automate.

For Your Business

If you're still using AI the way most people used it in 2023 — generating social posts, rewriting emails, maybe summarizing a document here and there — you're not wrong. But you're leaving the real value on the table. The question isn't "what can AI write for me?" It's "where in my business am I losing time, money, or information because a process is manual, inconsistent, or invisible?" That's where AI actually moves the needle.

If I could sit down with February 2023 Fran — maybe over a cup of Earl Grey — I'd tell her two things.

First: you're right to build tools. Keep doing that. But build them closer to the money. A quote generator saves someone twenty minutes. A dispatch system that recovers missed estimates saves them thousands. Both use the exact same instinct — see a friction point, build something that removes it — but the impact is on a completely different scale.

Second: write more. That blog post was 300 words. You had more to say. You always have more to say. The fact that you were already skeptical of AI-generated content in the very first weeks of ChatGPT's existence — that perspective is worth documenting in detail, not in a quick hit.

Your experience is the asset. The tools are just the medium.

Your experience is the asset. The tools are just the medium.

That February 2023 post is a time capsule. It captures the exact moment when the world was trying to figure out what this thing was, and I was already doing what I always do — trying to make it useful for the people I serve. Small business owners who need practical solutions, not hype. People who are running real companies with real employees and don't have time for the AI thought-leader circus.

Three years later, the tools are unrecognizable. The principle hasn't moved an inch.

Start with the person. Start with their problem. Build something that actually helps.

The rest is just technology.