What would a "counterculture of AI" look like?

Shitty math kills: shitty math "proved" that being selfish produced optimal outcomes and torched the planet; shitty math rains hellfire missiles down on innocents; in the 1960s, shitty math drove the "hamlet pacification program," producing 90,000 pages a month in data about the endless hail of bombs the US dropped on Vietnam.

The Spirit of 68 -- a wave of rebellions and uprisings around the world -- rallied around a rejection of shitty math and the first stirrings of a digital counterculture, based around seizing the means of computation.

Today, AI's shitty math is in some ways just more of the same, but it brings something new to the table: a deep bias for herding humans into behaving the same way, forever, and to behave as much like each other as is possible.

Dan McQuillan's piece on "AI dissidence" rehearses many of the well-known failings of AI and the well-publicized tech-worker resistance to weaponizing AI to unleash further harms, but then it gets into some pretty interesting new territory: asking what a "counterculture of AI" would look like.

He proposes that it will be "ludic" (playful), "playfully serious" (like the Situationists), and "fun, yet anti-fascist."

It's a fun idea to play with, one I'm going to sit with for a while. I think that McQuillan dismisses the "ethical AI" movement way too blithely; organizations like AI Now are led by people who are whip-smart, deeply ethical, and every bit as skeptical of capitalism and Big Tech any you might hope. Read the rest