tailcalled

Non-copy­a­bil­ity as a se­cu­rity feature

AI de­vel­op­ment as the first fully-au­to­mated job

Against Infrabayesianism

Knock­ing Down My AI Op­ti­mist Strawman

My Men­tal Model of AI Op­ti­mist Opinions

Evolu­tion’s se­lec­tion tar­get de­pends on your weighting

Em­pa­thy/​Sys­tem­iz­ing Quo­tient is a poor/​bi­ased model for the autism/​sex link

Bi­nary en­cod­ing as a sim­ple ex­plicit con­struc­tion for superposition

Ra­tion­al­ist Gnosticism

RLHF is the worst pos­si­ble thing done when fac­ing the al­ign­ment problem

[Question] Does life ac­tu­ally lo­cally *in­crease* en­tropy?

Why I’m bear­ish on mechanis­tic in­ter­pretabil­ity: the shards are not in the network

In defense of tech­nolog­i­cal un­em­ploy­ment as the main AI concern

The causal back­bone conjecture

Ra­tion­al­ists are miss­ing a core piece for agent-like struc­ture (en­ergy vs in­for­ma­tion over­load)

[LDSL#6] When is quan­tifi­ca­tion needed, and when is it hard?

[LDSL#5] Com­par­i­son and mag­ni­tude/​diminishment

[LDSL#4] Root cause anal­y­sis ver­sus effect size estimation

[LDSL#3] In­for­ma­tion-ori­en­ta­tion is in ten­sion with mag­ni­tude-orientation

[LDSL#2] La­tent vari­able mod­els, net­work mod­els, and lin­ear diffu­sion of sparse lognormals