Concept Extrapolation: A Conceptual Primer
This is a foundational discussion for AI safety researchers, focusing on preventing misalignment as AI systems evolve, but it is incremental as it builds on existing alignment concepts without new empirical results.
The article addresses the problem of model splintering, where concepts shift over time, by introducing concept extrapolation as a method to safely extend concepts to broader contexts, arguing it is essential for AI alignment.
This article is a primer on concept extrapolation - the ability to take a concept, a feature, or a goal that is defined in one context and extrapolate it safely to a more general context. Concept extrapolation aims to solve model splintering - a ubiquitous occurrence wherein the features or concepts shift as the world changes over time. Through discussing value splintering and value extrapolation the article argues that concept extrapolation is necessary for Artificial Intelligence alignment.