Jots ✎

I noticed that if I create a first version of text/feature/whatever with AI, with the idea of not having to start with a "blank page", that I find it suprisingly hard to engage mentally with it after that, to think through it properly.

It sounds like it makes sense, but I am not sure what the real reason actually is.
Is it emotional detachment because it was produced by AI, producing some kind of subconcious resistance on my side?
Is it that the output looks just good enough so I feel like I am wasting time / nitpicking by going through it, even though I know it is not where I want it to be?
Is it that seeing a single solution in front of me, my mind can't "de-focus" from it and allow itself to research other branches of the solution space?
Probably it is a combo of these.

I also noticed this with solutions presented by others: often, when they heavily used AI for initial prototyping, and solution is lacking, it is hard to get them to commit to thinking through it deeper, not because they don't think it is needed, but because it is suddenly a much bigger task than it would be if they did it from scratch themselves, because whole part of thinking, with all its parts of exploring the space, learning, throwing away possible solutions, building a final solution on top of all that, never happened.

Me, what I do now, when I care about the problem I am solving, I start from "blank page" and let myself think through it before I engage AI. I might use it to help with smaller steps and research bits during the way, but I make sure not to delegate thinking itself till I do a certain satisfying amount of it myself: in AI speak, I make sure I am the main agent and AI is only serving as sub-agents. Then, once I have thorough footing, I will let AI take on bigger parts, possibly offer its own solution next to mine, and so on. But from that point on, I don't find it hard anymore to engage with whatever AI will bring to me.

Being up to date with the latest innovations in the field gives you an edge to come up with an uniquely innovative idea. But then, the moment you start implementing it, you start falling behind in keeping up to date with the latest innovations.

You are trading width for depth.

That's why you always have a chance to come up with something better (in evolving fields), despite the incumbents: you can go wide in the way they can't. But then, once you pick your "fight" and go more in depth, you become open to later disruption.

There is nothing wrong about this, it is how things work. I guess you do need to try to balance both, keep some of that width while going into depth. But the best bet is really going into depth so hard, winning that one so much, that newcomers can't catch up with you or you have enough resources built up to pivot effectively when needed.