Designing Proactive Thought Partners for Writing
Writing involves diverse cognitive activities, from ideation to revision, and writers’ needs vary across individuals and moments. Proactive AI promises to provide the right support at the right time, yet existing proactive tools largely focus on generic textual assistance, such as autocomplete. This paper studies the design space of proactive thought partners: AI agents that proactively offer customizable, higher-level cognitive support during writing. We instantiated this concept in a technology probe and deployed it with 16 participants for one week. The probe allows users to create partners by configuring their roles and proactivity. As users write, relevant partners take the initiative at appropriate moments to offer suggestions. Our findings show that participants configured proactive support through prospective planning, used suggestions for both idea generation and self-monitoring, and valued lightweight visual representations alongside non-directive rhetorical framing for non-intrusive interventions. We derive implications for designing proactive writing assistants around customization, timing, engagement, and representation.
Introduction. Writing is a dynamic cognitive process (Flower and Hayes, 1981). From one moment to the next, a writer might shift between developing ideas, selecting prose, and evaluating written text in light of broader narrative goals. The support they need can therefore change throughout the writing process. At one moment, a writer may benefit from an example that makes an abstract claim concrete; at another, they may need help considering an alternative perspective or deciding how to develop the next section. These needs also vary across writers. Some writers may want support that helps them develop and evaluate their own ideas, while others may prefer information-seeking assistance, such as suggesting evidence from outside their domains. An ideal writing assistant should therefore support this shifting and personal cognitive work by offering the right kind of help at the right moment. Existing proactive writing assistance approaches this ideal in a narrower setting.
Discussion / Conclusion. In this paper, we explored proactive thought partners, a design concept that combines system initiative with customizable, cognitive support for writing. Drawing on findings from our probe study, we discuss design implications for future proactive writing support across four dimensions: customization, timing, engagement, and representation (Table 4). 10. Conclusion
Lines of inquiry this paper opens 24
Research framings built by reading the notes related to this paper — the questions it feeds into.
How do evaluation biases undermine LLM quality assessment systems?- How do LLMs generate false citations that sound like real scholarship?
- Why does automated evaluation consistently overestimate research quality?
- Can statistical filtering plus narrative generation fool academic peer review?
- How do retrieval failures enable generation of fabricated scholarly constructs?
- How do citation patterns encode collective judgment about research quality?
- What makes counterfeiting social warrant different from counterfeiting factual claims?
- Can discourse-level analysis detect deception better than individual word choices alone?
- How do verification labels themselves become part of the misinformation problem?
- Why does peer review fail on unrepeatable AI-generated outputs?
- Can verification mechanisms prevent AI agents from inventing false citations?
- Can we verify fabricated text without redesigning the generation process?
- Can human researchers verify automated research methods before they become uninterpretable?
- Can citation practices work when AI cannot produce traceable sources?
- How does treating synthetic data as empirical evidence contaminate statistical inference?
- What safeguards prevent AI from generating fake papers with fabricated citations?
- What happens when you reverse-engineer raw materials from published papers?
- What happens when lawyers rely on AI citations that turn out false?