Designing Proactive Thought Partners for Writing

Abstract

Writing involves diverse cognitive activities, such as ideation and revision, and writers’ needs vary across individuals and moments. Proactive AI promises to provide the right cognitive support at the right time, yet existing proactive tools largely focus on local textual assistance, such as autocomplete. This paper studies proactive thought partners: customizable AI agents that proactively offer higher-level cognitive support during writing. We instantiated this concept in a technology probe and deployed it with 16 participants over one week. The probe allows users to create partners by configuring their roles and timing. As users write, it monitors the editor and activates relevant partners to generate suggestions when support may be useful. Our findings show that customization became a form of prospective planning, while effective proactivity depended on contextual alignment, respect for current intentions, and preserved user control. We derive implications for the design space of proactive writing assistants around customization, timing, engagement, and representation.