If Tetra writes [a] Lesswrong post(s) about alignment she is proud of in 2024, which subfield(s) will it be about?
AI️ AI AlignmentTetraspace
➕
Plus
15
Ṁ1.6k
Jan 1
62%
Re-explaining the basics
48%
Decision theory
46%
Corrigibility
44%
Public communications
40%
Claude
37%
Embedded agency
26%
Policy
26%
Evals
25%
AI control (https://www.lesswrong.com/posts/kcKrE9mzEHrdqtDpE/the-case-for-ensuring-that-powerful-ais-are-controlled)
25%
Agent foundations
25%
Forecasting
22%
Mechanistic interpretability

Resolves YES to every alignment subfield that I write a LessWrong post that I'm proud of in 2024. Posts can be about multiple subfields, and all subfields will resolve YES. "Proud of" will resolve according to the same rules as the market on whether such a post exists.

Get Ṁ1,000 play money
Comments
15 Holders
47 Trades

Related questions