OpenAI Blog · May 3, 2018
AI safety via debate
Reviewed by Errol Vogt, Site support technician & online learning analyst · original summary · editorial policy
AI safety via debate. We’re proposing an AI safety technique which trains agents to debate topics with one another, using a human to judge who wins. This update is relevant for small-office operators tracking changes in their tools.
Operator takeaway: For operators: review whether 'AI safety via debate' affects your current setup before relying on it in production.
ai