back

by pr337h4m·3y ago·view on hn ↗
I've been convinced for sometime now that "effective altruism" (Lesswrong-ism? Yudkowskianism?) is the communism of the 21st century.

Lots of parallels here beyond the obvious common thread of utilitarian collectivism:

- Censorship: the concept of "infohazards" is super-popular https://nickbostrom.com/information-hazards.pdf https://www.lesswrong.com/tag/information-hazards

- Social control + the creation of an 'other': They are absolutely convinced that sentient AGI will inevitably kill everyone without any real evidence. However, if we manage to create sentient AGI, what these guys plan to do with it is best described as modern slavery https://www.lesswrong.com/tag/ai-boxing-containment#:~:text=...

- Surveillance: They justify it using Bostrom's "vulnerable world hypothesis": https://nickbostrom.com/papers/vulnerable.pdf?platform=hoots... (TFA also discusses it well)

- Desire for government power: TFA is super clear about that. More interestingly, here's the FTX Future Fund on space governance: "We’d love to see workshops or a mock constitutional convention where sharp people think hard about how to structure international governance institutions for the long-term future, or how to govern space settlement. As explained in more detail on our areas of interest page, we believe that the onset of space settlement could be a watershed moment in human history. We want people to start thinking about how it should work." https://ftxfuturefund.org/projects/alternative-voting-system...

Also: https://80000hours.org/problem-profiles/space-governance/

- Entryism/influence operations: Pretty much every career profile on 80,000 Hours encourages taking a job at government agencies/influential private organizations to push the EA agenda https://80000hours.org/career-reviews/

Also, https://en.wikipedia.org/wiki/Roko%27s_basilisk. No one with morals and a spine should fear this. (FWIW, the same justification of fearing retribution by an evil power for refusing to collaborate could be used by people in occupied countries who collaborated with the Nazis.)

2 comments
The base model Roko's basilisk is pretty stupid, but I have talked to people who have worked on "improving" the idea, so it actually works in more realistic and diverse decision theories. It's like base model MOND vs improved versions of MOND, the idea is still pretty wacky but they did some fudging to get it to work better "empirically" (with that word being used very loosely, of course). I think some people don't want anyone talking about the improved basilisks, but if anyone cares I can try to dig them up. Probably don't care though, because it's really stupid.
>I've been convinced for sometime now that "effective altruism" (Lesswrong-ism? Yudkowskianism?) is the communism of the 21st century.

I mean, maybe sorta.. the eschatology is far more interesting with EA, at least.

The EA eschatology is pretty much Christian eschatology, except they aren't at all sure their savior (a Friendly, Strong AI) will come. There's a good chance of just getting the Beast and the Harlot instead.

Hence the obsession with AI safety - Christians just have to work to bring Christ's kingdom to where they are. The EAs have to make sure nobody summons a vengeful demon overlord while they're slowly crafting their own all-loving, all-powerful digital deity.