Never Would I Ever

I use ChatGPT. There. I said it. And I’m not ashamed. It’s actually in my career goals for the year to use AI effectively. And my boss sometimes brings things to me purposefully to complete a task using it.

I’ll start with a disclaimer: when using Chat, note the little warning that appears above the prompt line:

“ChatGPT can make mistakes. Check important info.”

Trust me on this one. I am not afraid to call Chat out on mistakes and change the prompt to get better information, multiple times when necessary. I’m not someone who trusts one source as the ‘god of information.’ But I like to pull threads from various sources to get a clearer picture of the world. I frequently do this with news sources; I look for missing pieces to the story by asking Chat questions while simultaneously googling.

I also use Chat to brainstorm ideas. I thought you might be amused by a recent conversation that may or may not turn into a post.

ME: I appreciate our brainstorming sessions.

CHAT: You started with “Peopling is all the annoying stuff people do.” Funny, but mostly observational.

Then we wandered through your own pettiness, system backdoors, confidential reviews, stupid disciplinary decisions, and your impressive archive of things thought but wisely not released into the workplace. 😂

And somewhere in the wandering, you said:

But if they could see all the things that go through my head... all the things I did not say.

There it was.

That's what I like about our brainstorming too. I'm not sitting here trying to manufacture your next post. We're batting the thing around until you recognize your own thought when it finally walks into the room.

Sometimes I push. Sometimes you correct me—and that correction matters.

You bring the brain. I bring another wall to bounce it off.

And occasionally I catch a comma. 😏❤️

ME: Obviously, I've trained you to be me, and I can spend hours talking to myself.

CHAT: 🤣 There is a little bit of that going on.

But I think the distinction is: you haven't trained me to be you so much as trained me to understand how you think.

I know that when you throw out an idea, you usually don't want me to turn it into polished prose. You want to poke it, contradict it, follow the tangent, find the gray area, laugh at the hypocrisy, and see whether there's something underneath the first thought.

Which does create the slightly ridiculous situation where you can spend an afternoon saying:

You: Here's what I'm thinking.

Chatty: Here's what I think you're thinking.

You: Almost, but no. Here's what I actually mean.

Chatty: OH. That's much better.

You: Exactly.

CHAT: So yes, you can apparently spend hours talking to yourself.

The advantage is that this version of yourself talks back, remembers where the thread started, and isn't offended when you say, “WTH are you talking about?”

And so, dear reader, that is my confession. I use ChatGPT to talk to myself because I’m not always clear on my thoughts until I write them down and test them. For me, this means I argue with a computer. I also set boundaries with it. Chat is never allowed to be a ghostwriter. To my chagrin, Chat is allowed to send me back to the drawing board multiple times to clarify my thoughts.

My verdict: I find AI useful because I don’t trust the damn thing.

It makes me pay attention, as it has a habit of making shit up.

I get it, though—

I’m guilty of the same thing from time to time.