If Redwood Research releases an ELK benchmark paper, will I think it's great backchained empirical alignment research?
8
70
Ṁ485Ṁ130
resolved Sep 6
Resolved
YES1D
1W
1M
ALL
Resolved according to my subjective judgement.
This is roughly as described here: https://www.lesswrong.com/posts/DwoPGM8ytBCXrZpM7/backchaining-in-strategy
By 'great backchained empirical alignment research', I both mean the research is good and that it benefited from carefully reasoning through the alignment problem and possible approaches. This particularly includes thinking about the future situation we'll be in during high risk periods.
Resolves to n/a if no such paper is released in 1 year.
Get Ṁ1,000 play money
🏅 Top traders
# | Name | Total profit |
---|---|---|
1 | Ṁ61 | |
2 | Ṁ48 | |
3 | Ṁ20 | |
4 | Ṁ10 | |
5 | Ṁ5 |
More related questions
Related questions
Will someone strengthen our Goodhart's Law result?
14% chance
Will Superalignment succeed? (self assessment)
21% chance
Will someone give a correct explanation of what this is?
70% chance
Is the "100% effective against solid tumors" cancer pill AOH1996 paper legit? [see description]
23% chance