Back when I was in university, there were always a few fellow students who did a dumb thing: They would work like crazy to get out of having to do work. They would have all these elaborate plans to find loopholes, cheat, or save time, but doing so also cost them a lot of time. I always thought: “Why don't you just do the damn work?”
I see a similar thing happening with AI. But this time I'm falling into the same trap myself. For example, the other day, instead of entering a form on a website, I got Claude to do it. But then I had to spend so much time prompting Claude through it that I probably could have entered the data myself in the same time (if not less). I immediately thought: “Why didn't I just do the damn work?”
With AI, there's always an effort-to-payoff ratio. When you use it to automate a repetitive task—as David and I talk about all the time on Intentional AI—you may save substantial time in the long run. But for smaller tasks, the returns are far smaller. Sometimes they're negative.
This is a tried-and-true productivity principle: The purpose of productivity advice is to increase how much you accomplish per unit of time.
The next time you use Claude, ChatGPT, or Perplexity to get some work done, reflect: how much time did you save?
The best productivity tools allow us to make back our time—and then some.
