ResearcharXiv cs.AI
Constructive Alignment: Governing Preference Dynamics in Human-AI Interaction
#ai alignment#human preferences#interaction#dynamic systems#behavioral economics
English
The paper introduces the concept of Constructive Alignment, which redefines AI alignment as a dynamic process of governing human preferences rather than merely satisfying static preferences. It emphasizes that human preferences evolve through interaction with AI systems, and alignment should focus on regulating how these systems influence preference development over time.
中文
本文提出了建设性对齐的概念，将人工智能对齐重新定义为治理人类偏好的动态过程，而不仅仅是满足静态偏好。它强调人类偏好通过与人工智能系统的互动而发展，对齐应侧重于调节这些系统如何影响偏好的发展。