Designing for an AI-Augmented Journaling Experience: Balancing Guidance and Autonomy for Deeper Emotional Insight
Kevin Wang, Samantha Barg, Ramon Lawrence
Abstract
Journaling has long been recognized as a valuable avenue for selfreflection and emotional well-being, yet sustaining meaningful engagement can be difficult.In this paper, we explore the possibilities and pitfalls of integrating LLM-driven features, emotional dissection, semantic retrieval, and conversational prompting, into a personal informatics-focused journaling system.Through an autobiographical design approach, we iteratively designed with a prototype that leverages large language models and reflective informatics principles, probing how AI and data might enrich users' introspection without undermining their autonomy.Our findings demonstrate the benefits of holistic tracking and visualization in reinforcing consistent journaling habits, while also uncovering key tensions between automated guidance and self-directed reflection.By interlacing personal informatics with AI-driven prompts, we reveal new opportunities for enhancing emotional awareness, fostering sustained journaling practices, and uncovering long-term patterns.Additionally, we highlight privacy concerns and ethical considerations in handling sensitive personal data under the context of AI.We conclude by suggesting directions for future research in AI-supported personal informatics and journaling tools.