

This makes a lot of sense. There was a lot of validation in creating something or solving a problem that could only be achieved through focus and determination. But importantly, something that also had an actual solution. A lot of my work was about doing something hard that would bear fruits like a nice map, some new insight, or an elegant analysis.
Now most of what remains are problems that cannot really be answered through focus, but just infinite layers of social reality and philosophical problems.
Maybe it could be interesting to ask if it could be possible to find a new set of problems that are more similar to the data work of before.

When reading your comment, i had to think of this blog post that i read this morning: https://tante.cc/2026/11/17/on-moving-goalposts/
And that people act as if they validated the output of the machine, but if they would have actually done that properly, there would be not efficiency gain (in temporal terms) any more.
So the efficiency gain is doing things worse, as you describe it.