Looking for adversarial collaborators to test our Debate protocol

post by Beth Barnes (beth-barnes) · 2020-08-19T03:15:26.732Z · LW · GW · 5 comments

EDIT: We're also looking for people to become trained Honest debaters, which requires a greater time commitment (ideally >=5 hours per week for >= 2 months) but for which we're offering $30/hr. If you're interested in doing that, please fill out this form: https://forms.gle/2bv1Z8eCYPfyqxRF9


We’re looking for people to help us adversarially test our Debate [LW · GW] protocol.

We have a set of questions, a (somewhat complicated) set of rules and mechanisms for how the debate should work, a (slightly janky) web interface for conducting the debates, and a protocol for judging: we have a pipeline for selecting good judges through MTurk, and they get 5 minutes to judge the final round from the debate.

We think that the person who gives the better answer to the question at the start (the “honest debater”) should win the debate, if they understand why that answer is good, and they have practiced the “honest debater strategy” a bit.

We’re looking for people to play the ‘dishonest debater’ role, and win against our trained honest debaters.

We’re ideally looking for people who:

More details, rules, experiment plan and tips for debaters are here.

First we want to pilot our overall protocol with only a small number of adversarial collaborators, and we’ll probably find some holes in our experiment rules and general setup that are unrelated to the properties of the debate mechanism itself.

If we manage to fix the holes in the experimental protocol, but don’t believe we’ve found problems in the actual debate mechanism yet, we’ll probably try to escalate the adversarialness, for example by scaling it up to a larger number of people, or offering prizes for dishonest wins.

If you're interested, comment below or email me: barnes [at] openai.com.

If you would be interested in participating conditional on us offering pay or prizes, that's also useful to know.


Comments sorted by top scores.

comment by Ben Pace (Benito) · 2020-08-19T19:44:03.981Z · LW(p) · GW(p)
  • Believe that there’s a dishonest strategy that should win in these debates

This is one of my favorite job requirements ever.

comment by rohinmshah · 2020-08-19T18:58:07.481Z · LW(p) · GW(p)

Planned summary for the Alignment Newsletter:

OpenAI is looking for people to help test their <@debate@>(@Writeup: Progress on AI Safety via Debate@) protocol, to find weaknesses that allow a dishonest strategy to win such debates.
comment by Pattern · 2020-08-19T16:55:33.975Z · LW(p) · GW(p)

If you would be interested in participating conditional on us offering pay or prizes, that's also useful to know.

Do you want this feedback at the same address?

Replies from: beth-barnes
comment by Beth Barnes (beth-barnes) · 2020-08-19T18:24:08.185Z · LW(p) · GW(p)

Yep, or in comments. Thanks!

comment by habryka (habryka4) · 2020-08-19T03:44:22.818Z · LW(p) · GW(p)

This sounds fun! I probably won't have enough time to participate, but I do wish I had enough time.