LESSWRONG
LW

Leveling Up: advice & resources for junior alignment researchers

Feb 15, 2023 by Orpheus16

This sequence is a compilation of resources that I recommend to people who are leveling up in alignment research. I expect it to be most useful for junior alignment researchers (~0-2 years of experience).

This sequence is not a comprehensive set of resources. It's a set of resources that was cherry-picked by me. It also mostly focuses on my own writing, with a few strong pieces by others. The pieces focus on frames, tools, "ways of thinking", and general resources-- not content knowledge (for that, see AGISF & Alignment 201).

"The hope is that you and others like you will help actually solve the problem, not just follow directions or read what’s already been written." -- Abram Demski

1777 traps that (we think) new alignment researchers often fall into
Orpheus16, Thomas Larsen
3y
10
88Qualities that alignment mentors value in junior researchers
Orpheus16
2y
14
122Principles for Alignment/Agency Projects
Ω
johnswentworth
3y
Ω
20
284Alignment Research Field Guide
Ω
abramdemski
6y
Ω
11
167Worst-case thinking in AI alignment
Ω
Buck
4y
Ω
18
70Resources that (I think) new alignment researchers should know about
Orpheus16
3y
9
5011 heuristics for choosing (alignment) research projects
Orpheus16, danesherbs
2y
5
98Naive Hypotheses on AI Alignment
Shoshannah Tekofsky
3y
29
70Alignment Org Cheat Sheet
Orpheus16, Thomas Larsen
3y
8
34An overview of some promising work by junior alignment researchers
Orpheus16
3y
0
413(My understanding of) What Everyone in Technical Alignment is Doing and Why
Ω
Thomas Larsen, elifland
3y
Ω
90