Beyond Prompts: Learning from Human Communication for Enhanced AI Intent Alignment

TL;DR

This paper investigates human communication strategies for specifying intent to improve AI alignment, aiming to design AI systems that better understand and fulfill user goals by learning from human-human interactions.

Contribution

It introduces a novel approach by analyzing human communication strategies to enhance AI intent understanding and alignment, bridging human-human and human-LLM interactions.

Findings

Identifies key human communication strategies for intent specification

Proposes design principles for AI systems based on human communication

Enhances AI interpretability and user alignment through human-inspired methods

Abstract

AI intent alignment, ensuring that AI produces outcomes as intended by users, is a critical challenge in human-AI interaction. The emergence of generative AI, including LLMs, has intensified the significance of this problem, as interactions increasingly involve users specifying desired results for AI systems. In order to support better AI intent alignment, we aim to explore human strategies for intent specification in human-human communication. By studying and comparing human-human and human-LLM communication, we identify key strategies that can be applied to the design of AI systems that are more effective at understanding and aligning with user intent. This study aims to advance toward a human-centered AI system by bringing together human communication strategies for the design of AI systems.