[long] Some tests of how much AI "understands" what it says (spoiler: very little)
First, an apology for how fucking long this ended up being, in part thanks to how long winded AI responses are. David wanted me to post it here so I'm posting....
lmao it finally happened
Someone I was following on TikTok, whose takes on tech industry bullshit and specifically AI hype I respected, made a video that Roko's basilisk is a serious concern. My apologies to those who have been in this same situation when I was less sympathetic.
Chris Langan and the "Cognitive Theoretic Model of the Universe"? Oh boy!
a lesswrong: 47-minute read extolling the ambition and insights of Christopher Langan's "CTMU"...
Rationalist org bets random substack poster $100K that he can't disprove their covid lab leak hypothesis, you'll never guess what happens next
rootclaim appears to be yet another group of people who, having stumbled upon the idea of the Bayes rule as a good enough alternative to critical thinking, decided to try their luck in becoming a Serious and Important Arbiter of Truth in a Post-Mainstream-Journalism World....
I'm imagining you right now
I'm doing it so skillfully. I'm totally basilisking you. isn't that scary