LLM assisted journaling helps writers keep control and connect past reflections
AuthScribe: Supporting Agency and Authorship in LLM-Assisted Reflective Journaling
Human-Computer Interaction
Summary
Writing about personal experiences can help people understand themselves better. The authors created a tool called AuthScribe that uses AI to help people write reflective journals without losing their own voice or control. People using AuthScribe changed how they connected new experiences to their past thoughts over two weeks. They felt in charge when they could change AI suggestions into their own words and trusted prompts linked to their previous writings.
What this means in practice
- •For mental health app developers: Create journaling apps that help users keep personal authorship while leveraging AI suggestions for self-reflection.$Commercial implications: Enables personalized, privacy-preserving AI journaling apps that maintain user control, appealing to wellness product markets.
- •For productivity tool designers: Build writing tools that integrate AI support allowing users to revise AI suggestions into meaningful content and track personal context over time.
Authors
Lanping Hu, Yinsi Zhou, Zhenchang Xing, Xiaoyu Sun, Shidong Pan
Abstract
Large language models are increasingly used in reflective journaling, where articulating an experience is part of interpreting it. This raises questions about how writers retain authorship, exercise agency, and assess personalized assistance. We present AuthScribe, an LLM-assisted reflective journaling system that combines contextual anchors, writer-steered AI support, Autobiographical Citations, and read-only Provisional Drafts that require re-expression before AI-generated suggestions enter the journal. Retrieval and LLM inference run on-device. In a 14-day field deployment with 14 participants, grounding shifted from situating current experiences to reconnecting them with prior self-authored interpretations. Perceived agency remained stable through Week 1 and shifted during Week 2. Participants described authorship through transforming AI-generated suggestions into personally meaningful expression, and trusted personalized prompts when they could trace them to prior writing. We discuss design implications for adoption friction, longitudinal agency, and traceable personalization in AI systems for personal reflection.