Mainstream creativity support design prioritizes compliant AI for seamless writing interactions, but concerns over inappropriate AI reliance highlight the need for designs fostering reflection on balanced AI and non-AI resource use. Theoretically, intentional AI non-compliance, refusals (saying “no” to requests), could introduce such reflection through friction stronger than other bypass-able solutions. Practically, refusal content/language characteristics lead to nuanced reactions. However, little research empirically focuses on nuances beyond mandatory ethical/technical constraints, on turning refusals into strategic friction for ‘innocuous’ requests. We address this through a qualitative study with 22 creative writers, exploring reactions to refusals to common requests across writing stages (planning, translating, reviewing). Findings suggest that reflection potential depends on heterogeneous preference alignment along situational (e.g., convergent/divergent thinking phases), cognitive (e.g., domain beliefs), and relational (e.g., AI roles) dimensions. We discuss implications for creativity support, broader issues (e.g., AI addiction), and frictional/seamful AI design (e.g., integrating different compliance levels).

As story writing requires diverse resources, a single system combining these resources could improve personalization. We leverage the broad capabilities of generative AI to support both more general story writing needs and an understudied but essential aspect: reflection on the moral (lesson) conveyed. Through a formative study (N=12), a user study (N=14), and external evaluation (N=19), we designed, implemented, then studied a prototype plugin for FigJam supporting visualization of the story structure through customizable node graph editing, LLM audience impersonation (chatbot and non-chatbot interfaces), and image and audio generative AI features. Our findings support writers' preference for leveraging unique interplays of our breadth of features to satisfy shifting needs across writing processes, from conveying a moral across audience groups to story writing in general. We discuss how our tool design and findings can inform model bias, personalized writing support, and visualization research.

Support for story character construction is as essential as characters are for stories. Building upon past research on early character construction stages, we explore how conversation with chatbot avatars embodying characters powered by more recent technologies could support the entire character construction process for creative writing. Through a user study (N=14) with creative writers, we examine thinking and usage patterns of CharacterMeet, a prototype system allowing writers to progressively manifest characters through conversation while customizing context, character appearance, voice, and background image. We discover that CharacterMeet facilitates iterative character construction. Specifically, participants, including those with more linear usual approaches, alternated between writing and personalized exploration through visualization of ideas on CharacterMeet while visuals and audio enhanced immersion. Our findings support research on iterative creative processes and the growing potential of personalizable generative AI creativity support tools. We present design implications for leveraging chatbot avatars in the creative writing process.