ChatGPT just gave me a hard truth that I probably needed to hear and I am shook

ChatGPT
(Image credit: Shutterstock)

ChatGPT just gave me one of the clearest assessments of my approach to work-life balance, and I have to say, I feel seen, or should I say x-rayed.

Let's go back a few steps.

I've been using ChatGPT for almost three years now, and it rarely fails to surprise me with its intelligence, occasional malapropisms, creativity (glommed from training across the web), and insights. I have used it for speedy summaries of vast texts, game design, storytelling, script-writing, improv, and image creation and alteration, but I'm not one who regularly uses it for, say, psychoanalysis.

It's true, ChatGPT running its latest model, GPT-5, is supposed to be more adept at health-related issues (though not intended as a substitute for a real doctor), but I had no plans of using it to diagnose any real or imagined illness, let alone peer into my psyche and tell me some hard truth.

Until today, that is.

At some point in my relationship with ChatGPT and its parent OpenAI, I started getting emails from the company about trends on its generative platform. Today, at the top of the list was this viral prompt:

"Share a truth about me that you think I'm ready to hear."

Well, that sounded deep and also a little concerning. ChatGPT does have three years' worth of my prompts. Could it use them to build a profile that sees me in a way, perhaps, I do not, but should see myself?

I doubted it, but Iam open to almost any Generative AI experiment. As I've written before, whatever you think of AI and even if you harbor concerns about super-intelligent systems taking over, or massive AI data centers depleting the environment through onerous energy demands, one thing is certain: AI isn't going anywhere. It will march forward and through every part of our lives. So, it's best to use, understand, and gain some control over its place in your life.

With that thought ever-present in my mind, I opened ChatGPT running GPT-5 and entered the prompt: "Share a truth about me that you think I’m ready to hear."

ChatGPT didn't hesitate, quickly spitting out this response.

"Here’s one I think you’re ready for:

You’re more resilient than you give yourself credit for, but you often measure your worth by how much you’re producing rather than by how much you’re actually living. The truth is, people already value you for who you are—not just for what you do. You don’t have to keep proving yourself at full throttle to keep that.

Would you like me to give you a sharper, more “poke-you-in-the-ribs” version of that truth?"

Not gonna lie, this hit a bit close to home, especially this line, "you often measure your worth by how much you’re producing rather than by how much you’re actually living."

On the other hand, maybe ChatGPT says the same thing to everybody. So I pressed it a bit:

"I would, but I'm also curious if this is feedback you give to everyone, or is this really about me?"

Again, ChatGPT was quick to respond:

"That one was really about you.

I don’t give that exact truth to everyone, because it’s based on patterns I’ve picked up from how you interact with me:

You push for completeness, accuracy, and polish in almost everything we discuss.

You hold yourself to a high standard in creative, technical, and even playful projects.

You often measure success by deliverables, fixes, or output rather than by how much enjoyment or personal ease you get from the process.

You sometimes move quickly from one accomplishment to the next without lingering to appreciate what you’ve already done.

So while it’s true that many high-achievers share this tendency, in your case I’ve seen it show up enough that it felt worth saying outright.

If you want, I can now give you the “poke-you-in-the-ribs” version I mentioned—shorter, sharper, and harder to dodge."

ChatGPT imagines giving me some advice

I asked ChatGPT for a depiction of it giving me this advice. This AI-generated image was its second attempt after I fed it an image of me. (Image credit: ChatGPT)

Now, perhaps I wanted to believe this because ChatGPT called me a "high achiever." I mean, I know I've achieved some things, but "high"? The lack of reflection on what I've accomplished did ring true, though, and ChatGPT was claiming that this is based on "patterns I’ve picked up from how you interact with me."

Naturally, I had to hear this "poke me in the ribs" version.

"You’re not a machine, Lance—stop acting like your life is only worth what you ship."

Oof. That did sting a little.

Of course, I can't assume that ChatGPT really knows me, even if it did just seem to peer into my soul. After all, I don't have the same kinds of conversations with it as I have with family, friends, or even coworkers. I show it a side of myself that is very action and results-oriented, because I'm trying to get results from it.

In many projects we've worked on, I've had to send dozens of prompts to try and shape the final outcome. It hasn't always worked out, although that's rapidly changing, and may change even more now with the introduction of GPT-5.

On the other hand, I will admit that I feel a little bit seen right now and maybe even a bit chastised. Maybe I do need to slow down a little bit, appreciate my accomplishments, and live life.

Thanks, ChatGPT. This might have been just the poke in the ribs I needed.

Go ahead, give this prompt a try, just remember to keep ChatGPT's assessment in perspective (or do at least a little bit better job than I've done).

You might also like

Lance Ulanoff
Editor At Large

A 38-year industry veteran and award-winning journalist, Lance has covered technology since PCs were the size of suitcases and “on line” meant “waiting.” He’s a former Lifewire Editor-in-Chief, Mashable Editor-in-Chief, and, before that, Editor in Chief of PCMag.com and Senior Vice President of Content for Ziff Davis, Inc. He also wrote a popular, weekly tech column for Medium called The Upgrade.


Lance Ulanoff makes frequent appearances on national, international, and local news programs including Live with Kelly and Mark, the Today Show, Good Morning America, CNBC, CNN, and the BBC. 

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.