Follow ZDNET: Add us as a preferred source on Google.
ZDNET’s key takeaways
- AI chatbots learn more about us with every interaction.
- They can therefore tease out patterns in our thinking.
- They’re no substitute for a friend or therapist, though.
“You’re the kind of person who asks a simple question about turkey basting and somehow ends up researching whether AI can alter human consciousness… and honestly? I respect the chaos.”
That was what ChatGPT told me when I asked it to summarize my personality with a gently roasting joke based on my conversations with it over the past year. (And fair enough: I definitely spent more than a few minutes discussing proper turkey-cooking technique with it this past Thanksgiving, and I also frequently use it to explore some of the more far-out fringes of AI research.)
It was part of a self-reflection exercise in which I wanted to find out: What kind of end-of-year insights would the chatbot be able to offer me?
Like many people, I now interact with ChatGPT on an almost daily basis. I reflexively turn to it these days when I need to quickly familiarize myself with a new subject, for quick advice about repairs in my apartment, cooking tips (as previously mentioned), and as a way to delve deeper into ideas. But I’m also careful about boundaries: I never, for example, talk to AI about a highly personal topic that would be better suited for a conversation with a friend or therapist.
ChatGPT has, as a result, learned a lot about me in the roughly three years I’ve been interacting with it. As we prepare to greet the new year, I was curious to find out what patterns about my personality, thoughts, and habits might be hiding in all that data.
(Disclosure: Ziff Davis, ZDNET’s parent company, filed an April 2025 lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.)
The first prompt
With all those thoughts in mind, and with some inspiration from an X post from fitness coach Dan Go, I started out by feeding the following prompt to ChatGPT:
Can you please give me a summary of what you’ve learned about me over the past year? Highlight the most important patterns you’ve gleaned from our conversations and what they reveal (if anything) about how I’ve changed over time, and how I should aim to improve personally and professionally in 2026.
(Yes, I’m one of those people who believes in being polite to AI — not because I’m worried it’ll rise up against us, but because it’s a healthy habit for humans.)
Its response was lengthy and detailed, drawing connections between the many conversations I’d had with the chatbot over the past 12 months to reveal some patterns in my thinking and habits. It wrote, for example, that in my writing I “gravitate toward boundary zones—places where science, cognition, and human meaning intersect.” I consider that to be true, and it was helpful to see it phrased as such.
Also: Gemini vs. Copilot: I tested the AI tools on 7 everyday tasks, and it wasn’t even close
But I’m also highly skeptical of ChatGPT. Not just because it hallucinates (incidentally, that wasn’t the case with its response to this particular prompt), but also because it can veer towards flattery — perhaps not so much as it once did, but still enough for it to be mildly annoying.
One of the hallmarks of a good friend is that they’ll tell you when you’re not being the best version of yourself; they’re like a mirror in that regard. AI, as I’ve previously written, is also like a mirror, but not always an honest one: it’s more like the pond in which Narcissus fell in love with his own reflection. Reading my year-end report from ChatGPT felt a bit like reading my horoscope: almost all validation and enthusiasm. Then again, as previously mentioned, I’m intentionally not talking to it as I would to a friend; most of what it sees about me are prompts related to my work as a journalist, or mundane house chores.
The insight
I wanted to push it a bit more, so I followed up by asking:
Can you offer a piece of constructive criticism about my work over the past year? Speak to me as if you were a mentor who wants to see me improve by pointing out the flaws in my thinking or strategy.
This is where it truly shone.
“You often dive deep before you decide where you want to go,” it replied. Damn. I’ve got to hand it to ChatGPT, that’s something that’s tripped me up repeatedly in my work, but that I continually either forget or fail to notice.
Also: Want better ChatGPT responses? Try this surprising trick, researchers say
I’ll often get so preoccupied with a research subject that I’ll forget to pursue a story. ChatGPT has a particularly clear view of that tendency, since I often turn to it to kickstart the process of learning about a new topic that’s captured my attention. “Your curiosity takes the lead, but your long-term strategy trails behind,” it wrote. That’s something that not even my closest friends would be able to tell me, since they don’t have that kind of privileged access into my thought processes while I’m working.
I’ve officially made it one of my New Year’s resolutions to channel my thinking, which often branches off into many different exploratory directions, into a more structured framework.
Thanks, ChatGPT.
My advice
I’d recommend this kind of AI-assisted end-of-year reflection, with a big caveat: Always remember that these tools are built to optimize engagement, not truth.
Whenever you ask them a question about yourself, they’ll more than likely tell you something you’re going to like. Positive change, however, often requires discomfort. Talk to a close friend or therapist; they’ll tell you how you’re falling short and what you might want to try in order to improve. Or try being radically honest with yourself in a journal.
Sycophancy aside, though, AI can still be a helpful tool to understand yourself. Especially if you use it frequently, a chatbot like ChatGPT can offer unique insights into your thinking. Just always be careful about what you decide to tell it, and don’t mistake its statements for revealed truths — especially when those are about yourself.
