TY - GEN
T1 - "Help Me Help the AI"
T2 - 2023 CHI Conference on Human Factors in Computing Systems, CHI 2023
AU - Kim, Sunnie S.Y.
AU - Watkins, Elizabeth Anne
AU - Russakovsky, Olga
AU - Fong, Ruth
AU - Monroy-Hernández, Andrés
N1 - Publisher Copyright:
© 2023 Owner/Author.
PY - 2023/4/19
Y1 - 2023/4/19
N2 - Despite the proliferation of explainable AI (XAI) methods, little is understood about end-users' explainability needs and behaviors around XAI explanations. To address this gap and contribute to understanding how explainability can support human-AI interaction, we conducted a mixed-methods study with 20 end-users of a real-world AI application, the Merlin bird identification app, and inquired about their XAI needs, uses, and perceptions. We found that participants desire practically useful information that can improve their collaboration with the AI, more so than technical system details. Relatedly, participants intended to use XAI explanations for various purposes beyond understanding the AI's outputs: calibrating trust, improving their task skills, changing their behavior to supply better inputs to the AI, and giving constructive feedback to developers. Finally, among existing XAI approaches, participants preferred part-based explanations that resemble human reasoning and explanations. We discuss the implications of our findings and provide recommendations for future XAI design.
AB - Despite the proliferation of explainable AI (XAI) methods, little is understood about end-users' explainability needs and behaviors around XAI explanations. To address this gap and contribute to understanding how explainability can support human-AI interaction, we conducted a mixed-methods study with 20 end-users of a real-world AI application, the Merlin bird identification app, and inquired about their XAI needs, uses, and perceptions. We found that participants desire practically useful information that can improve their collaboration with the AI, more so than technical system details. Relatedly, participants intended to use XAI explanations for various purposes beyond understanding the AI's outputs: calibrating trust, improving their task skills, changing their behavior to supply better inputs to the AI, and giving constructive feedback to developers. Finally, among existing XAI approaches, participants preferred part-based explanations that resemble human reasoning and explanations. We discuss the implications of our findings and provide recommendations for future XAI design.
KW - Explainable AI (XAI)
KW - Human-AI Collaboration
KW - Human-AI Interaction
KW - Human-Centered XAI
KW - Interpretability
KW - Local Explanations
KW - XAI for Computer Vision
UR - http://www.scopus.com/inward/record.url?scp=85152299586&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85152299586&partnerID=8YFLogxK
U2 - 10.1145/3544548.3581001
DO - 10.1145/3544548.3581001
M3 - Conference contribution
AN - SCOPUS:85152299586
T3 - Conference on Human Factors in Computing Systems - Proceedings
BT - CHI 2023 - Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems
PB - Association for Computing Machinery
Y2 - 23 April 2023 through 28 April 2023
ER -