Kaj_Sotala(Kaj Sotala)

[Question] How to talk about rea­sons why AGI might not be near?

Step­ping down as mod­er­a­tor on LW

How I ap­ply (so-called) Non-Violent Communication

Most peo­ple should prob­a­bly feel safe most of the time

A brief col­lec­tion of Hin­ton’s re­cent com­ments on AGI risk

Ro­mance, mi­s­un­der­stand­ing, so­cial stances, and the hu­man LLM

Good­hart’s Law in­side the hu­man mind

[Question] Why no ma­jor LLMs with mem­ory?

Creat­ing a fam­ily with GPT-4

Here, have a calm­ness video

[Fic­tion] The boy in the glass dome

The Prefer­ence Fulfill­ment Hypothesis

In Defense of Chat­bot Romance

Fake qual­ities of mind

Jack Clark on the re­al­ities of AI policy

Open & Wel­come Thread—July 2022

My cur­rent take on In­ter­nal Fam­ily Sys­tems “parts”

Con­fused why a “ca­pa­bil­ities re­search is good for al­ign­ment progress” po­si­tion isn’t dis­cussed more

The hor­ror of what must, yet can­not, be true

[In­visi­ble Net­works] Goblin Marketplace