AI Safety via Debate

ESRogs5 May 2018 2:11 UTC

LW: 27 AF: 11

New paper and blog post by Geoffrey Irving, Paul Christiano, and Dario Amodei (the OpenAI safety team).

What links here?

Some Thoughts on Metaphilosophy by Wei Dai (10 Feb 2019 0:28 UTC; 76 points)
Debate Minus Factored Cognition by abramdemski (29 Dec 2020 22:59 UTC; 37 points)
The “AI Debate” Debate by michaelcohen (2 Jul 2020 10:16 UTC; 20 points)
Running your own workshop on handling hostile disagreements by Camille Berger (8 Nov 2023 10:28 UTC; 12 points)
Wei Dai's comment on Best reasons for pessimism about impact of impact measures? by TurnTrout (10 Apr 2019 20:03 UTC; 10 points)
Wei Dai's comment on Raemon’s Shortform by Raemon (12 Aug 2019 7:42 UTC; 4 points)
ESRogs's comment on Thoughts on “AI safety via debate” by Gordon Seidoh Worley (10 May 2018 3:36 UTC; 3 points)
ESRogs's comment on Thoughts on AI Safety via Debate by Vaniver (10 May 2018 3:35 UTC; 2 points)
ESRogs's comment on Understanding Iterated Distillation and Amplification: Claims and Oversight by William_S (5 May 2018 2:13 UTC; 2 points)

ESRogs5 May 2018 2:11 UTC

LW: 27 AF: 11