What is interesting to me is that thought leaders are always optimising for some X - be it productivity, some higher good, low poverty, efficiency etc. But two

What is interesting to me is that thought leaders are always optimising for some X - be it productivity, some higher good, low poverty, efficiency etc. But two things humanity, as a whole, consistently optimises for is *short-term survival* and *lesser day-to-day friction*. ChatGPT/AI seems to be just another friction optimisation, and truth is, we don't know if this optimisation leads us to the world that is anywhere close to 'pretty' by modern standards. People from pre-industrial world would be shocked to see modern people enjoying all the air and noise pollution, but there is no way to avoid falling into such local optimum, because we do not optimise for clean air or quality of life, only for lower friction and survival. This seems to put us randomly into some local optimums all the time, until unsolved problems pile up so high that they start endangering our survival - rinse and repeat. What I'd argue is there is no way known to mankind that allows to steer this globally in a different way.