Personalised language models influence user trust and sharing habits
Tailored to you: longitudinal effects of personalising language models
Artificial Intelligence
Summary
People use personalised language models in different ways depending on how the models remember or learn about them. The study found that repeated use of any language model changes how people interact with it, but personalisation can lead to more or less sharing of personal details and feelings about privacy. For instance, models that remember past conversations make users share more and feel less creeped out. On the other hand, models personalised from upfront surveys can make users regret sharing personal info. The authors highlight that different ways of personalising affect users in complicated ways.
What this means in practice
- •For ai product designers: Design AI assistants that adapt memory-based personalisation to encourage user trust and comfortable self-disclosure over time in daily interactions.$Commercial implications: This paper enables developing consumer AI assistants tailored to individual communication patterns, improving user experience and engagement.
- •For privacy engineers: Create privacy safeguards for AI systems using survey-based personalisation to prevent inducing user regret about data sharing.
Authors
Canfer Akbulut, Justine Breuch, Arianna Manzini, Lujain Ibrahim, Matija Franklin, Roma Patel, Iason Gabriel, Kristian Lum, Laura Weidinger
Abstract
Interest in developing personalised language models is rapidly growing. While personalisation is often viewed as a mechanism to better serve diverse user needs, the effects of sustained interactions with personalised models on people's perception of and behaviour toward AI remain poorly understood. Most critically, downstream consequences outside the immediate human--AI interaction loop, such as effects on users' self-perceptions and interpersonal relationships, remain largely unexamined. In this study, we recruited 992 participants to complete daily advice-seeking interactions with language models over the course of five days, comparing outcomes from a non-personalised baseline against two personalisation approaches: memory-based (conditioned on prior conversational history) and survey-based (conditioned on information collected through a pre-study intake survey). We find that several changes in human-AI interaction over time are driven primarily by repeated exposure rather than personalisation itself. However, participants interacting with personalised models experienced differences in advice-seeking and information-sharing attitudes and behaviours: participants in the memory-based condition engaged in greater self-disclosure and rated the model as less creepy, while participants in the survey-based condition reported higher regret about having shared personal information with the AI. We conclude by highlighting the nuanced effects of different personalisation approaches on interaction outcomes, and discussing the implications of these findings for the responsible design and deployment of personalised AI systems.