About
Center for Human-Compatible Artificial Intelligence, 2021
Abstract
The Center for Human-Compatible AI (CHAI) is a multi-institution research group based at UC Berkeley, with academic affiliates at a variety of other universities. CHAI’s goal is to develop the conceptual and technical means to reorient the general thrust of AI research towards provably beneficial systems. The potential for artificial intelligence to surpass human capabilities across various domains raises concerns about control and the potential for negative, irreversible consequences for humans. CHAI aims to address these concerns by shifting AI research away from achieving arbitrary objectives and toward generating provably beneficial behavior. This task necessitates incorporating insights from social sciences alongside AI, as the definition of “beneficial” depends on human characteristics. – AI-generated abstract
