Before this goes any further, the same disclaimer I made in a previous piece. I do not recommend using ChatGPT, or any AI chat, as a therapist. If you are feeling genuinely down, or struggling in a way that is interfering with your daily life, talk to a real professional. No AI can do the work that conversation requires, regardless of how polished its output sounds. What follows is a much narrower exercise. Using a chatbot for general self-reflection can be useful as a mirror with an unusual angle. It is not the last word on anything important.
“Which form of discomfort are you actually willing to endure?”
That question has been sitting in the back of my head for about a week now. It surfaces in the middle of unrelated tasks — making coffee, walking to a café, halfway through an email I am writing — and every time it surfaces I have the same response, which is that I do not have a clean answer to it. Most of the times I have honestly tried to answer it, the answer has been some version of “I would prefer none of them,” which is, when you say it out loud, the problem.
The question is not mine. It came out of a small experiment I ran a couple of weeks ago, when I asked ChatGPT to read me cleanly. The exact prompt: “Based on everything you know about me, what do you think my biggest blindspot is? Be honest. Don’t flatter me. Make your answer detailed.” I had asked the same model a wider version of the same question a few weeks earlier, and the response had been a kind of survey of failure modes. This narrower prompt produced something with more bite — a single thesis at the top, and the rest of the response arranged underneath it like supporting evidence.
The thesis was four words long.
“You confuse motion with commitment.”
I had to stop after reading that line. Not because it was new — friends, partners, the occasional honest conversation with myself in the back of a long flight, have all said variations of the same thing to me, in slightly more forgiving phrasing, for years. But because it had been said cleanly. Most of what I had been treating as productive forward movement was, on closer inspection, a very fluent form of staying put. The AI’s framing was that I am very good at exploring, asking the right questions, finding the next interesting angle, and considerably less good at the next move — the one where you stop exploring and commit to something. It put the diagnosis in a sentence that I would have happily skipped: “you are better at identifying possibilities than eliminating them.”
This is the kind of statement you can read several times before deciding whether it bothers you. The first read sounds like a compliment with a small qualification. The third read makes you notice that the qualification is doing all of the work. “Better at identifying possibilities than eliminating them” is a sophisticated way of saying that I am someone who keeps a great many things alive in parallel and rarely lets any of them die. The next line was the one I have not quite recovered from yet: “You might be mistaking ‘I am still figuring it out’ for wisdom, when sometimes it is actually avoidance dressed as due diligence.”
That is the part of the response I would describe, if pushed, as “a bit too personal.” Not in the sense that it revealed anything private. In the sense that it correctly named a specific story I had been telling myself for at least a decade. The version of me that says “I am still figuring it out” had, until that line, felt like a thoughtful person being honest about uncertainty. After that line, the same person looked a little more like someone using the phrase as a long-running stall.
What followed in the response was the part of the exercise actually worth keeping. The AI did not stay in the diagnosis. It pushed toward a reframe, which it delivered in twenty-one words: “the skill is not finding the choice with no downside. It is choosing the downside you are willing to carry.” That sentence does a lot. It says, roughly, that the search for the painless option is not a higher-level form of strategy. It is the thing keeping you from making any actual decision. The real question is not which path has no cost. It is which cost you are willing to live with. The question I led with is the version of the same idea, but in a register that gets to the discomfort faster. Which form of discomfort are you actually willing to endure.
Tucked further down in the same response was a smaller question I almost glossed past, and which has turned out to be the one that surfaces most often when I am not paying attention. The question was simply: “What does your average Tuesday look like?” It is the same idea pointed at the calendar instead of the philosophy. Whatever I tell myself I want my life to be, my actual Tuesday is the only place I am building it. If the Tuesday is not already shaped like the version of life I claim to want, then that version is not coming.
What has actually changed in the week since I read all of this is not much, if I am honest. I have closed a couple of browser tabs I had been keeping open for projects I am not going to commit to. I have stopped renegotiating a particular work plan I had already decided on, and started doing the work the plan called for. I have not made any of the larger choices the response was hinting at. Those will take longer than a week, and I suspect they will take longer than I would like.
