AI4SG Lab @ CHI 2026

🏅 Honourable Mention

AI-exhibited Personality Traits Can Shape Human Self-concept through Conversations [PDF]

Jingshu Li, Tianqi Song, Nattapat Boonprakong, Zicheng Zhu, Yitian Yang, Yi-Chieh Lee

Does chatting with an AI change who you think you are? We discovered that users unconsciously adopt a chatbot's personality traits during conversations, and surprisingly, we actually enjoy the chat more as we become more like the bot!

Understanding, Mitigating, and Leveraging Cognitive Biases to Calibrate Trust in Evolving AI Systems  [Homepage] [Proposal]

Saumya Pareek, Nattapat Boonprakong, Naja Kathrine Kollerup, Si Chen, Simo Hosio, Koji Yatani, Yi-Chieh Lee, Ujwal Gadiraju, Niels van Berkel, Jorge Goncalves

As LLMs become more persuasive, how do our own cognitive biases trick us into trusting them too much, or too little? Join our interactive 90-minute workshop to collaborate with cross-disciplinary researchers, map out open challenges, and shape the future research agenda for human-AI trust.

Can AI be a Social Buffer? Investigating the Effect of AI-assisted Cognitive Reappraisal and Narrative Perspectives on Managing Difficult Workplace Conversations over Email [PDF]

Chi-Lan Yang, Jing Li, Xuhui Chang, Jingshu Li, Koji Yatani, Yi-Chieh Lee

Ever receive a stressful workplace email and wish you had an emotional shield to soften the blow? We explored how using AI to positively reframe difficult messages helps receivers process bad news and significantly reduces negative emotions.

Fit Matters: Format–Distance Alignment Improves Conversational Search [PDF]

Yitian Yang, Yugin Tan, Jung-Tai King, Yang Chen Lin, Yi-Chieh Lee

Should an AI search assistant give you a high-level summary or concrete details with images? Come see how matching information formats to a user's 'psychological distance' reduces cognitive load and makes decision-making a breeze.

Exploring the Human-LLM Synergy in Advancing Theory-driven Qualitative Analysis

Han Meng, Yitian Yang, Wayne Fu, Jungup Lee, Yunan Li, Yi-Chieh Lee

LLMs are great at basic qualitative coding, but can they actually help you discover new theories? Meet CHALET, a novel human-LLM collaborative approach that uses iterative coding and 'disagreement analysis' to uncover deep, latent themes you might have missed.

🏅 Honourable Mention

Affective and Goal-Oriented Factors of Relationship Formation in the Digital Therapeutic Alliance: A Longitudinal Study of Mental Health Chatbots

Zian Xu, Yi-Chieh Lee, Karolina Stasiak, Jim Warren, Danielle Lottridge

Do users need to trust a mental health bot before they open up? Our research flips the script, showing that delivering practical and emotional support is actually what builds trust and satisfaction, not the other way around.

🏅 Honourable Mention

Who You Explain To Matters: Learning by Explaining to Conversational Agents with Different Pedagogical Roles [PDF]

Zhengtao Xu, Junti Zhang, Anthony Tang, Yi-Chieh Lee

Does teaching an AI help you learn better, or does a challenging AI push you further? Join us to explore how different pedagogical roles radically shift the dynamics and outcomes of human-AI educational interactions.

Understanding Older Adults' Experiences of Support, Concerns, and Risks from Kinship-Role AI-Generated Influencers [PDF]

Tianqi Song, Black Sun, Jingshu Li, Han Li, Chi-Lan Yang, Yijia Xu, Yi-Chieh Lee

What happens when older adults start scrolling through short-video apps and bonding with 'AI grandchildren'? We unpack how these virtual influencers are fulfilling real emotional needs while raising complex questions about the future of family ties.

🏅 Honourable Mention

Designing Computational Tools for Exploring Causal Relationships in Qualitative Data [PDF]

Han Meng, Qiuyuan Lyu, Peinuan Qin, Yitian Yang, Renwen Zhang, Wen-Chieh Lin, Yi-Chieh Lee

Can a computational tool map causal relationships in qualitative data without disrupting your established research workflow? Come see how QualCausal provides cognitive scaffolding for researchers while navigating the boundaries of traditional QDA practices.

AI Personalization Paradox: Reading Highlights for Personalized AI-Assisted Writing Increases Engagement but Undermines Autonomy and Ownership [PDF]

Peinuan Qin, Chi-Lan Yang, Nattapat Boonprakong, Jingzhu Chen, Yugin Tan, Yi-Chieh Lee

Personalized AI writing tools are supposed to empower us, but they might be doing the exact opposite. Come see how our study reveals the hidden risks of AI personalization, where users lose their sense of ownership and become overly reliant on the system.

Navigating Marginalization: Toward Justice-Oriented Sociotechnical Design for Parent–Child Learning among Southeast Asian Immigrant Mothers in Taiwan

Ying-Yu Chen, Yang Hong, Yan-Rong Chen, Yi-Chieh Lee

How do immigrant mothers pass on their cultural values when faced with structural marginalization? We unpack their creative home-learning strategies and propose a new playbook for designing systems that help diverse families flourish.

ChatLearn: Leveraging Non-Native Speaker Communication Challenges as Language Learning Opportunities [PDF]

Peinuan Qin, Yugin Tan, Jingzhu Chen, Nattapat Boonprakong, Zicheng Zhu, Naomi Yamashita, Yi-Chieh Lee

Stop letting AI do all the talking for you. Come see how ChatLearn transforms everyday multilingual chats into personalized learning opportunities, proving you don't have to sacrifice conversation flow to build your vocabulary.

When Humans Don’t Feel Like an Option: Contextual Factors That Shape When Older Adults Turn to Conversational AI for Emotional Support [PDF]

Mengqi Shi, Tianqi Song, Zicheng Zhu, Yi-Chieh Lee

Why would an older adult choose to confide in a chatbot instead of their own family? We dive into the hidden moments when seniors turn to AI for emotional support to preserve their dignity, save face, and avoid feeling like a burden.

InterPilot: Exploring the Design Space of AI-assisted Job Interview Support for HR Professionals [PDF]

Zhengtao Xu, Zimo Xia, Zicheng Zhu, Nattapat Boonprakong, Yu-An Chen, Rabih Zbib, Casimiro Pio Carrino, Yi-Chieh Lee

Can a real-time AI assistant actually make hiring interviews easier, or does it just add another screen to look at? Come see how our prototype, InterPilot, helps HR professionals juggle tasks, while uncovering the delicate design balance between AI assistance and human agency.

ConvScale: Conversational Interviews for Scale-Aligned Measurement

Peinuan Qin, Jingzhu Chen, Yitian Yang, Han Meng, Zicheng Zhu, Yi-Chieh Lee

Tired of subjecting your users to boring, rigid Likert scales? Meet ConvScale, an AI-powered approach that turns traditional psychometric questionnaires into natural, engaging conversations while still extracting accurate quantitative scores.

A Tree-Structured Interface for Efficient Revisitation in Long-Horizon LLM Conversations

Jiaming Li, Peinuan Qin, Yi-Chieh Lee

Ever get lost scrolling up and down a long AI chat trying to manage complex, multi-step tasks? Meet Branchat, a tree-structured interface that replaces the endless scroll, giving you total control over your context history and drastically reducing cognitive load.