luminous

@july articulates way more elegantly the issue which i've been trying to think through practically about AI. 

i call this meaningmaking: the work of deciding how valuable things are relative to other things. 

humans do this all the time. we do it when we decide that existing solutions aren't good enough, and when we make something new that is different and better. all innovation work is about meaningmaking, whether it is the invention of multitouch screens, national healthcare free at point of delivery, or a new narrative format (sebald ftw).

when we surrender meaningmaking to machines (including AI tools), we surrender the only work that is truly human to entities that can't do meaningmaking at all. the only correct way to use AI systems is to understand them as incredibly sophisticated tools, but as nothing more than tools.

we would not ask a knife what we should eat for dinner — but someone who learns how to use a knife well can make dinner better.

more here: https://vaughntan.org/meaningmakingai