Clojurians Log v2

To quote Claude: > He’s the entries’ truth in every read path — you’re right, that trade was imaginary.

It's a good read. Sounds like Gates too acknowledges that if we don't do anything, AI will run riot with the existing incentive structures in place (economic, geopolitical). That's the main takeaway for me from this piece.

I see lots of assertions that "AI will help with $THING" that are predicated on the assumption that AI will provide "nearly error-free work" and that is presented as "when" (not if)

A common issua circling AI is that too many sales pitches are taken for granted. As an engineer my first reaction is always "How do you know? What did you measure or study?"

With opinions we should always think about the possible motives. Especially if the opinions are all-encompassing, affect societys as a whole or are politically colored.

also I found surprising that Gates doesn't even mention accountability or liability issues in the glorious future where humans have been taken out of the loop and AI does all expert work

seems important to figure out, who is responsible for errors that create harm... there's a wide gulf between "nearly error-free" and actually error-free

Interesting read, if a bit heavy on inevitability. He also fails to mention the huge environmental impact.

but, you see, AI will give farmers "better advice"... so climate is not a problem

that part especially seemed to deliberately miss the mark