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

TL;DR

The paper tackles the problem of misalignment between user intent and AI outputs in the era of large language models. It conducts a within-subject comparative study of human-human versus human-LLM interactions to uncover effective intent-alignment strategies. Key findings show that human assistants proactively seek information, tailor responses to user history, and solicit feedback, while LLMs tend to respond passively and overwhelm users with broad option lists. The authors derive design implications for LLM-based interfaces, advocating proactive elicitation, recipient-design driven personalization, and enhanced multimodal communication to improve user alignment and experience.

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.