x
This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
LW
Login
Thomas Kwa's
Shortform
by
Thomas Kwa
22nd Mar 2020
AI Alignment Forum
1 min read
681
12
Ω 7
This is a special post for quick takes by
Thomas Kwa
. Only they can create top-level comments. Comments here also appear on the
Quick Takes page
and
All Posts page
.
Rendering 0/681 comments, sorted by
top scoring
(show all)
Click to highlight new comments since:
Today at 11:45 PM
Moderation Log
More from
Thomas Kwa
View more
Curated and popular this week
681
Comments
681
Thomas Kwa's Shortform — LessWrong
Mentioned in
172
"I am an AI Safety Researcher"
146
The world's best gradual disempowerment model organism: Frontier AI labs
81
Overtly misaligned trajectories score highly in RL.
80
Thomas Kwa's research journal
46
AI #75: Math is Easier
Load More (5/7)
Comment Permalink
Thomas Kwa
3y
*
3
0
Current posts in the pipeline:
Dialogue on whether agent foundations is valuable, with Alex Altair. I might not finish this.
Why junior AI safety researchers should go to ML conferences
Summary of ~50 interesting or safety-relevant papers from ICML and NeurIPS this year
More research log entries
A list of my mistakes from the last two years. For example, spending too much money
Several flowcharts / tech trees for various alignment agendas.
Reply
See in context