Every so often you get to a point where the same three or four questions have been circling for months and none of them are getting answered. The kind of questions you already know the shape of. Where to spend the next few years. What to keep and what to cut. What sort of life you are quietly building, whether you meant to or not.
I had been carrying a version of that quiet stack for a while when I opened ChatGPT and typed one prompt: “Based on everything you know about me, map out a five year plan. Don’t flatter me. Be detailed and realistic.”
The response was long and quite specific. It read like a plan a fairly perceptive stranger might write about me after a few hours of conversation. Some of it was familiar in a slightly uncomfortable way. Some of it landed in a different place than I expected.
Before going any further, I should be plain about how I use these tools. I reach for AI for ideas and for reflection — a way of thinking out loud, a mirror I can push against. It is not a replacement for actual thinking, and it is not a replacement for a real conversation with a trusted friend, a family member, or an actual professional whose job it is to help you (a therapist, a career coach, a doctor). The American Psychological Association has been clear on this: a chatbot is not equipped to do what a qualified person can do. I take that seriously, and I try not to confuse a well-worded paragraph on a screen for an outside voice that actually knows me.
With that on the table, here is what the exchange actually gave me.
The line that stayed with me longest was not about any of the practical stuff. It was this: “Freedom without structure often becomes procrastination with a nice view.” That is a sentence I could have written about my own last five years. I have chosen a version of life that is perhaps unusually free — I split my year between Ireland and Southeast Asia, I choose my own hours to a large extent. All of which is real. And all of which can quietly slide into a shape where each week looks a bit like the last one and it becomes hard to tell whether anything is being built.
The freedom is genuine. So is the drift, when I am not watching for it.
The second thing that landed was the point about health starting to become “a business asset” at this stage of life. Not the flattering health-optimization version — no cold plunges or supplement stacks — just the quieter observation that energy, sleep, a back that still works, a decent walk on most days, and moderate drinking are the actual inputs into being able to do good work. I already play golf, work out at home, and walk when I can. But I noticed how easily I can treat those things as bonuses rather than the base layer. When they slip, everything downstream slips too, and I usually only notice a couple of weeks later.
The third one was the “boring rhythm” note. That freedom, again, needs a schedule to hold it up — three days of deep work, a day for editing and admin, a day off, a flexible day. I don’t run my week that neatly. Nobody I know runs their week that neatly. But the observation is right that without any external structure at all, every day quietly becomes half of a good day. The office cues that used to end a workday no longer exist for me. The end of the day now has to be a decision, not a signal, and I am not always great at making it.
Then there was the piece about saying no. It said the next five years should be about focus, ownership, and compounding — not more effort. Subtraction, not addition. I have written before that saying no is not one of my strengths. Reading it framed as the actual work of the next five years, rather than a minor personal weakness, made it feel more serious. The default question it offered — “will this still matter in two years?” — is a small enough test that I can actually run it. I have not been running it.
The part I sat with the longest was the least dramatic. It described a worst-case version of the next five years that was not failure. It was being very busy, publishing constantly, knowing a lot, earning okay money, and waking up at 40-something with no clear thing that I own. That is not a scary sentence. That is a plausible sentence. The scary version of a life is often not the collapse. It is the slow one, where the shape is fine on paper and nobody would tell you to worry.
What the exercise did was name a few things I have been carefully not naming. That is what I use these tools for at their best, and it is also where I have to be most honest with myself about their limits. A good response from a chatbot can sound like clarity, and clarity is one of the things it is easiest to confuse with actually having thought something through. A five-year plan produced in 40 seconds is not a five-year plan; it is a decent set of prompts for the harder conversations — with my wife, with people whose judgment I trust, with myself on a long walk with the phone in my pocket. If any of it is going to become a plan, those are the places it will happen.
