Modulating Language Model Experiences through Frictions

TL;DR

This paper introduces selective frictions inspired by behavioral science to modulate user interactions with language models, reducing over-reliance and misuse while maintaining accuracy, based on a user study in a question-answering context.

Contribution

It proposes a novel approach of applying behavioral-inspired frictions to improve human-AI interaction and demonstrates their effects through empirical user studies.

Findings

Friction reduces user over-reliance and click rates.

Friction minimally impacts accuracy on targeted topics.

Unintended behavioral effects observed beyond frictioned topics.

Abstract

Language models are transforming the ways that their users engage with the world. Despite impressive capabilities, over-consumption of language model outputs risks propagating unchecked errors in the short-term and damaging human capabilities for critical thinking in the long-term. How can we develop scaffolding around language models to curate more appropriate use? We propose selective frictions for language model experiences, inspired by behavioral science interventions, to dampen misuse. Frictions involve small modifications to a user's experience, e.g., the addition of a button impeding model access and reminding a user of their expertise relative to the model. Through a user study with real humans, we observe shifts in user behavior from the imposition of a friction over LLMs in the context of a multi-topic question-answering task as a representative task that people may use LLMs for, e.g., in education and information retrieval. We find that frictions modulate over-reliance by driving down users' click rates while minimally affecting accuracy for those topics. Yet, frictions may have unintended effects. We find marked differences in users' click behaviors even on topics where frictions were not provisioned. Our contributions motivate further study of human-AI behavioral interaction to inform more effective and appropriate LLM use.