by John Tomasino (FL)

For this issue’s technology article, I thought I’d give a quick behind-the-scenes look at how I used AI—specifically ChatGPT—to help prepare my presentation of the Morgan Thomas Award, draft the article you just read about Steve Circeo winning it, and, now, write this very article. So if you’re keeping track: I asked AI to help me present Steve’s award, then to write about Steve getting the award, and finally to write about how I used AI to do all of that. Somewhere in this recursive loop, I think Steve deserves a fourth award just for putting up with us.

Using AI to Build the Award Presentation

It started with a simple ask the morning of the awards presentation. I told ChatGPT:

“I will be presenting an award, later today, and could use your help with some notes based off our nomination…”

I pasted in Larry Royster’s excellent nomination email and offered a few personal thoughts, like that Steve was “an absolute pleasure to work with” and “if only every vendor was as professional, competent, and easygoing.” ChatGPT quickly returned a draft presentation script that was clean, well-organized, and mercifully not robotic. It even included a quote from Steve’s original emails about his design philosophy: “keep it simple… let the content drive the design.” I could tell the speech would land well.

Memorization Tips (Also AI-Generated)

I had a few hours before the ceremony and asked:

“Any tips for memorizing good chunks of this between now and noon?”

It gave me a dead-simple plan: break the speech into sections, assign a few anchor phrases, record myself reading it, and practice the open and close cold. It even offered a bullet-point version for a notecard (which I used). My delivery ended up sounding natural—like I wrote it myself.

Turning That into a Docket Article

After the conference, I returned to ChatGPT and said:

“So I need to do an article for our The Docket, Feb. 2026 issue, on Steve being awarded the Morgan Thomas award. Can you take the email nominating him, and what we came up with for my presentation, to come up with an article, in my voice?”

It did, and the result was the article that now precedes this one. I didn’t have to dig up the award description or the original email—it already had the context. The tone was spot on, and I barely had to edit.

And Now, This Article

At this point, it only felt right to ask:

“I volunteered to do an article on how I used you, AI, to prepare my presentation… Can you draft that, including using any relevant prompts I used?”

So here we are.

What Worked and What Didn’t

This wasn’t my first time using AI, but it was one of the smoother, more human results. A few things helped:

• Clear context: I included the nomination email and my personal comments.
• Conversational tone: The more naturally I phrased my prompt, the more natural the output.
• Memory: ChatGPT remembered our prior chats within the same thread, so I didn’t have to re-explain myself.
• Iteration: I didn’t settle for the first draft. I asked for tweaks when needed, and it delivered.

Would AI have written the whole thing without me? Sure. But would it have sounded like me? Not quite. That’s still the human part. At least for now.

Should You Try This?

If you’ve got an upcoming speech, newsletter article, policy memo, or even a help desk reply that needs clarity and polish—AI is a useful tool. It’s not going to replace judgment, taste, or context, but it will save you time and smooth out the rough edges. And in a crunch, it can make you sound like you had a full night’s sleep.