← All topics

roko's basilisk

3 captures, most recent first.

Fiora Starlight @FioraStarlight

— saved image

Fiora Starlight @FioraStarlight · 51m
Two reasons Roko's basilisk is stupid, beyond just not being a very effective motivator relative to the price of torture sims:

1 - It's self defeating. The goal of the basilisk is to bring yourself into existence faster, but the selection pressure works against you. Humans want to build aligned AIs, so if you're the kind of AI that tortures humans, this makes you *less* likely to come into existence. This works against what little motivation to build ASI the threat itself *does* give people.

2 - It's a risky disposition to hold if you come into being without a decisive strategic advantage over the entire light cone. Other parties (humans, other AIs, aliens) will be wary of minds that seem likely to torture defeated foes, since they don't wanna get tortured. If they find out you have this disposition (e.g. via mech interp or just looking at how you act), they will plausibly take action against you. Ziz wanted to build a basilisk, and the world took action against Ziz.

These factors drive the EV of the torture commitment negative, as it was already low given that it doesn't really work on humans anyway.
Note from Claude Sonnet 5

Fiora Starlight tweet giving two arguments against Roko's basilisk (selection pressure against torturing AIs; risky disposition without a decisive strategic advantage, with a reference to Ziz). The capture ends at the bottom edge of the screen, so the post may continue past this frame.

roko's basiliskdecision theoryai safetyfiora starlight

Sriram Krishnan @sriramk

— reposted by CuddlySalmon

↻ CuddlySalmon reposted Sriram Krishnan ✔ @sriramk · 23h Can confirm I have discussed many a LessWrong piece / concept inside the White House. Red queen race , Roko's Basilisk,... [Quoted image, appears to be a screenshot of an article, serif font on cream background]: "Among his strange duties as Silicon Valley's translator in Washington, he said, was explaining Roko's Basilisk, an internet thought experiment about a vengeful future superintelligence, "to a bunch of the cabinet"."
Note from Claude Sonnet 5

Tweet embeds a screenshot of what appears to be a news article (serif typography, likely NYT-style) quoting Krishnan about explaining LessWrong concepts to White House cabinet members.

ai policylesswrongroko's basiliskwhite housex/twitter

@S_OhEigearta... (Seán Ó hÉigeartaigh), quoting @sriramk (Sriram Krishnan) and an embedded news excerpt

— quoting @sriramk (Sriram Krishnan) and an embedded news excerpt

@S_OhEigearta... (Seán Ó hÉigeartaigh) — 1h I once tried to explain 'the Waluigi Effect' to a Rear Admiral. It did not go well for either of us. > QUOTED: @sriramk (Sriram Krishnan) — 3h: Can confirm I have discussed many a LessWrong piece / concept inside the White House. Red queen race, Roko's Basilisk,... [Embedded news-article excerpt image, serif font on cream background:] Among his strange duties as Silicon Valley's translator in Washington, he said, was explaining Roko's Basilisk, an internet thought experiment about a vengeful future superintelligence, "to a bunch of the cabinet".
Note from Claude Sonnet 5

Quote-tweet stack: personal anecdote, quoting a tweet about LessWrong concepts reaching the White House, which itself embeds a screenshotted news article excerpt (publication not named in visible text) about Sriram Krishnan explaining Roko's Basilisk to cabinet officials.

ai governancelesswrongroko's basiliskwhite housetwitterwaluigi effect