Conversational ai interfaces with multiple personas reduce bias and thinking shortcuts
Beyond the Monolithic Interface: A Research Agenda for Plural Conversational AI
Human-Computer InteractionComputers and Society
Summary
Most AIs today talk to people using just one voice or personality, which can make us think less, believe less widely, and lose out on different cultural or value perspectives. The authors point out these problems and suggest a new design where multiple AI personas talk together at the same time. This way, users hear different viewpoints, which helps avoid narrowing their beliefs and encourages more active thinking. The idea comes from game design and other fields and could lead to better AI conversations.
What this means in practice
- •For chatbot developers: Design conversational agents using multiple AI personas simultaneously to provide users with diverse viewpoints and reduce bias and mental shortcuts.
- •For user experience designers: Create multi-persona conversational interfaces to enhance user engagement by presenting varied perspectives rooted in game design principles.
A position paper. It proposes an approach and reports no results.
Authors
Avijit Ghosh, Afifah Kashif, Isabelle Kohout, Zhenzhen Qi
Abstract
The current paradigm of artificial intelligence (AI) design elevates conversational chatbot interfaces, making them the default mode through which hundreds of millions of people interact with AI systems globally. This narrative review identifies three related harms produced by the single-voice chatbot interface, drawing on insights from literature in cognitive science, political studies, AI alignment, game design and human computer interaction. Firstly, cognitive offloading of user thinking, secondly, ideological narrowing of user beliefs, and thirdly, epistemic compression of cultural and value diversity. We then propose multi-persona interfaces, in which multiple personas are presented to the user simultaneously, as a cost-effective intervention that addresses all three harms at once. We ground this intervention in traditions the AI literature has largely overlooked, drawing on techniques from game design and participatory simulation, including the magic circle, role differentiation, and persona embodiment. We propose a taxonomy for the future design of multi-persona interfaces, and call for a research agenda that closes the gap between technical pluralistic alignment methods and empirical studies of human-AI interactions with pluralistic interfaces.