>types of problems AI excels at are generally those that humans have found but overlooked.

Seems to be a deep and interesting angle lurking here, especially when coupled to

>evolutionary approach will slow down

Meaning something I will have to think about (while keeping Graeber in the background[0]) or even plan around :)

Will respond after I sleep on it

[0] https://davidgraeber.org/articles/value-as-the-importance-of...

In value terms, the question becomes: who has the right to translate their money into what sorts of meaning? Who controls the medium through which, and the institutions through which, our actions become meaningful to ourselves, by the very act of being publicly recognized in some kind of public arena? It seems to me that while if one is trying to understand the strategies by which people can move back and forth between “fields”, and especially, by which some are excluded from them, Bourdieu’s models are pretty much indispensable, THEY DO LITTLE to tell us why anyone wishes to enter certain fields to begin with.

The evolution of fields, of domains of enquiries. AI-independent

First of all, I take it for granted there is really no such thing as “intelligence”

Writing like this from 2005, feels like a personal bedtime Bible for AI age