Goodhart's Law

Applied to Resolutions to the Challenge of Resolving Forecasts by Noosphere89 at 1mo

Applied to Soft optimization makes the value target bigger by Vladimir_Nesov at 1mo

Applied to Don't align agents to evaluations of plans by TurnTrout at 2mo

Applied to Don't design agents which exploit adversarial inputs by TurnTrout at 3mo

Applied to Alignment allows "nonrobust" decision-influences and doesn't require robust grading by TurnTrout at 3mo

Applied to Scaling Laws for Reward Model Overoptimization by leogao at 4mo

Applied to Outer alignment and imitative amplification by Noosphere89 at 4mo

Applied to Oversight Leagues: The Training Game as a Feature by Paul Bricman at 5mo

Applied to Can "Reward Economics" solve AI Alignment? by Q Home at 5mo

Applied to Reducing Goodhart: Announcement, Executive Summary by Ruby at 6mo

Applied to The Dark Miracle of Optics by Noosphere89 at 6mo

Applied to Bayesianism versus conservatism versus Goodhart by Noosphere89 at 6mo

Applied to Circumventing interpretability: How to defeat mind-readers by Lee Sharkey at 7mo

Applied to Proxy misspecification and the capabilities vs. value learning race by Ruby at 9mo

Applied to Goodhart's Law Causal Diagrams by Raemon at 10mo

Applied to Replacing Karma with Good Heart Tokens (Worth $1!) by jimrandomh at 10mo

Applied to [Intro to brain-like-AGI safety] 10. The alignment problem by Steven Byrnes at 10mo

Applied to Practical everyday human strategizing by [anonymous] at 10mo

Applied to Why Agent Foundations? An Overly Abstract Explanation by plex at 10mo