LESSWRONG
LW

Curated Posts

Archive Recommendations

Spotlight Items

Some AI research areas and their relevance to existential safety
Best of LessWrong 2020

Andrew Critch lists several research areas that seem important to AI existential safety, and evaluates them for direct helpfulness, educational value, and neglect. Along the way, he argues that the main way he sees present-day technical research helping is by anticipating, legitimizing and fulfilling governance demands for AI technology that will arise later.

by Andrew_Critch
Load More
74
Proposal for making credible commitments to AIs.
Cleo Nardo
3h
22
148
X explains Z% of the variance in Y
Leon Lang
3d
23
217
Do Not Tile the Lightcone with Your Confused Ontology
Ω
Jan_Kulveit
6d
Ω
26
172
Futarchy's fundamental flaw
dynomight
9d
48
167
Estrogen: A trip report
cube_flipper
12d
41
81
A Straightforward Explanation of the Good Regulator Theorem
Alfred Harwood
17d
28
142
Broad-Spectrum Cancer Treatments
sarahconstantin
20d
10
150
The Best Reference Works for Every Subject
Parker Conley
24d
27
254
Truth or Dare
Duncan Sabien (Inactive)
1mo
52
206
Winning the power to lose
KatjaGrace
1mo
86
316
Interpretability Will Not Reliably Find Deceptive AI
Ω
Neel Nanda
1mo
Ω
66
532
Orienting Toward Wizard Power
johnswentworth
1mo
142
Load More
470Welcome to LessWrong!
Ruby, Raemon, RobertM, habryka
6y
74
944AGI Ruin: A List of Lethalities
Ω
Eliezer Yudkowsky
3y
Ω
711
902Where I agree and disagree with Eliezer
Ω
paulfchristiano
3y
Ω
224
865Eight Short Studies On Excuses
Scott Alexander
15y
253
842Preface
Eliezer Yudkowsky
10y
17
785The Best Textbooks on Every Subject
lukeprog
14y
416
680What an actually pessimistic containment strategy looks like
lc
3y
138
677SolidGoldMagikarp (plus, prompt generation)
Ω
Jessica Rumbelow, mwatkins
2y
Ω
206
660AI 2027: What Superintelligence Looks Like
Ω
Daniel Kokotajlo, Thomas Larsen, elifland, Scott Alexander, Jonas V, romeo
3mo
Ω
222
646Simulators
Ω
janus
3y
Ω
168