Update on Ought's experiments on factored evaluation of arguments

📖 Description

[Ought](https://ought.org/) has written a detailed update and analysis of recent experiments on factored cognition. These are experiments with human participants and don't involve any machine learning. The goal is to learn about the viability of [IDA](https://www.alignmentforum.org/s/EmDuGeRw749sD3GKd?_ga=2.44167719.2055190071.1578702594-1142780176.1552454685), [Debate](https://openai.com/blog/debate/), and related [approaches](https://medium.com/@deepmindsafetyresearch/scalable-agent-alignment-via-reward-modeling-bf4ab06dfd84) to AI alignment. For background, here are some prior LW posts on Ought: [Ought: Why it Matters and How to Help](https://www.lesswrong.com/posts/cpewqG3MjnKJpCr7E/ought-why-it-matters-and-ways-to-help)), [Factored Cognition presentation](https://www.lesswrong.com/posts/DFkGStzvj3jgXibFG/factored-cognition).

📊 Game Impacts

💭 Reactions

🔗 Sources

• [1] https://www.alignmentforum.org/posts/pH3eKEAEupx8c2ep9/update-on-ought-s-experiments-on-factored-evaluation-of