Lessons from Convergent Evolution for AI Alignment

Jan_Kulveit and rosehadshar

27 Mar 2023 16:25 UTC

33 points

5 comments8 min readLW link

A stylized dialogue on John Wentworth’s claims about markets and optimization

So8res25 Mar 2023 22:32 UTC

115 points

12 comments8 min readLW link

My Objections to “We’re All Gonna Die with Eliezer Yudkowsky”

Quintin Pope21 Mar 2023 0:06 UTC

318 points

173 comments33 min readLW link

$500 Bounty/Contest: Explain Infra-Bayes In The Language Of Game Theory

johnswentworth25 Mar 2023 17:29 UTC

77 points

6 comments2 min readLW link

Deep Deceptiveness

So8res21 Mar 2023 2:51 UTC

190 points

44 comments14 min readLW link

More information about the dangerous capability evaluations we did with GPT-4 and Claude.

Beth Barnes19 Mar 2023 0:25 UTC

219 points

43 comments8 min readLW link

(evals.alignment.org)

The Waluigi Effect (mega-post)

Cleo Nardo3 Mar 2023 3:22 UTC

575 points

169 comments16 min readLW link

Understanding and controlling a maze-solving policy network

TurnTrout, peligrietzer, Ulisse Mini, montemac and David Udell

11 Mar 2023 18:59 UTC

286 points

17 comments22 min readLW link

Truth and Advantage: Response to a draft of “AI safety seems hard to measure”

So8res22 Mar 2023 3:36 UTC

87 points

9 comments5 min readLW link

Natural Abstractions: Key claims, Theorems, and Critiques

LawrenceC, Leon Lang and Erik Jenner

16 Mar 2023 16:37 UTC

169 points

13 comments45 min readLW link

[Question] What happens with logical induction when...

Donald Hobson26 Mar 2023 18:31 UTC

18 points

2 comments1 min readLW link

Discussion with Nate Soares on a key alignment difficulty

HoldenKarnofsky13 Mar 2023 21:20 UTC

208 points

26 comments22 min readLW link

SolidGoldMagikarp (plus, prompt generation)

Jessica Rumbelow and mwatkins

5 Feb 2023 22:02 UTC

646 points

194 comments12 min readLW link

“Publish or Perish” (a quick note on why you should try to make your work legible to existing academic communities)

David Scott Krueger (formerly: capybaralet)18 Mar 2023 19:01 UTC

93 points

43 comments1 min readLW link

Towards understanding-based safety evaluations

evhub15 Mar 2023 18:18 UTC

121 points

9 comments5 min readLW link

Shell games

TsviBT19 Mar 2023 10:43 UTC

78 points

8 comments4 min readLW link

What Discovering Latent Knowledge Did and Did Not Find

Fabien Roger13 Mar 2023 19:29 UTC

135 points

10 comments11 min readLW link

Anthropic’s Core Views on AI Safety

Zac Hatfield-Dodds9 Mar 2023 16:55 UTC

178 points

38 comments2 min readLW link

(www.anthropic.com)

Descriptive vs. specifiable values

TsviBT26 Mar 2023 9:10 UTC

13 points

1 comment2 min readLW link

the QACI alignment plan: table of contents

carado21 Mar 2023 20:22 UTC

49 points

0 comments2 min readLW link

(carado.moe)