Better Slow than Sorry: Introducing Positive Friction for Reliable Dialogue Systems

TL;DR

Introducing positive friction into dialogue systems involves deliberately slowing conversations at strategic moments to promote reflection, improve goal alignment, and enhance task success, counteracting the risks of overly frictionless AI interactions.

Contribution

This paper proposes a novel framework of positive friction in dialogue systems, including an ontology and expert annotations, to improve accountability and understanding in goal-oriented AI interactions.

Findings

Positive friction fosters accountable decision-making.

It enhances machine understanding of user beliefs and goals.

Increases task success rates in experiments.

Abstract

While theories of discourse and cognitive science have long recognized the value of unhurried pacing, recent dialogue research tends to minimize friction in conversational systems. Yet, frictionless dialogue risks fostering uncritical reliance on AI outputs, which can obscure implicit assumptions and lead to unintended consequences. To meet this challenge, we propose integrating positive friction into conversational AI, which promotes user reflection on goals, critical thinking on system response, and subsequent re-conditioning of AI systems. We hypothesize systems can improve goal alignment, modeling of user mental states, and task success by deliberately slowing down conversations in strategic moments to ask questions, reveal assumptions, or pause. We present an ontology of positive friction and collect expert human annotations on multi-domain and embodied goal-oriented corpora. Experiments on these corpora, along with simulated interactions using state-of-the-art systems, suggest incorporating friction not only fosters accountable decision-making, but also enhances machine understanding of user beliefs and goals, and increases task success rates.