[Paper Review] From Human-Computer Interaction to Human-AI Interaction: New Challenges and Opportunities for Enabling Human-Centered AI.

Wei Xu, Marvin J. Dainoff+2 more|arXiv (Cornell University)|May 12, 2021

Human-Automation Interaction and Safety61 references22 citations

TL;DR

This paper proposes a human-centered AI (HCAI) framework grounded in human-AI interaction (HAII), emphasizing unique challenges in AI systems distinct from traditional HCI. By analyzing AI-specific issues and promoting interdisciplinary HAII research, the authors offer enhanced HCI methods and strategic recommendations to guide practitioners in designing ethical, user-centered AI systems.

ABSTRACT

While AI has benefited humans, it may also harm humans if not appropriately developed. We conducted a literature review of current related work in developing AI systems from an HCI perspective. Different from other approaches, our focus is on the unique characteristics of AI technology and the differences between non-AI computing systems and AI systems. We further elaborate on the human-centered AI (HCAI) approach that we proposed in 2019. Our review and analysis highlight unique issues in developing AI systems which HCI professionals have not encountered in non-AI computing systems. To further enable the implementation of HCAI, we promote the research and application of human-AI interaction (HAII) as an interdisciplinary collaboration. There are many opportunities for HCI professionals to play a key role to make unique contributions to the main HAII areas as we identified. To support future HCI practice in the HAII area, we also offer enhanced HCI methods and strategic recommendations. In conclusion, we believe that promoting the HAII research and application will further enable the implementation of HCAI, enabling HCI professionals to address the unique issues of AI systems and develop human-centered AI systems.

Motivation & Objective

• To identify unique challenges in AI systems that differ from non-AI computing systems, particularly in human-AI interaction.
• To advance the human-centered AI (HCAI) approach by focusing on the distinct characteristics of AI technology.
• To position HCI professionals as key contributors to HAII through interdisciplinary collaboration.
• To develop enhanced HCI methods and strategic recommendations tailored for AI system design.
• To support the practical implementation of HCAI by addressing AI-specific interaction issues.

Proposed method

• Conducting a comprehensive literature review focused on AI development from an HCI perspective.
• Comparing non-AI computing systems with AI systems to highlight unique interaction challenges.
• Extending the HCAI framework originally proposed in 2019 to address emerging AI-specific concerns.
• Proposing HAII as a new interdisciplinary domain to integrate HCI expertise into AI system development.
• Developing enhanced HCI methods and strategic recommendations for practical application in HAII.
• Emphasizing the role of HCI professionals in shaping ethical, user-centered AI through systematic methodological guidance.

Experimental results

Research questions

• RQ1 What unique challenges in AI systems are not adequately addressed by traditional HCI approaches?
• RQ2 How can the human-centered AI (HCAI) framework be further advanced through human-AI interaction (HAII) research?
• RQ3 In what ways can HCI professionals contribute meaningfully to the design and implementation of human-centered AI systems?
• RQ4 What enhanced HCI methods and strategic recommendations are needed to support HAII practice?
• RQ5 How can interdisciplinary collaboration in HAII enable more ethical and effective AI development?

Key findings

• AI systems introduce unique interaction challenges not present in non-AI computing systems, requiring new approaches beyond traditional HCI.
• The human-centered AI (HCAI) framework must be extended through dedicated human-AI interaction (HAII) research to address AI-specific issues.
• HCI professionals are well-positioned to contribute to HAII due to their expertise in user-centered design and interaction modeling.
• Enhanced HCI methods and strategic recommendations are essential for supporting the practical implementation of HCAI in real-world AI systems.
• Interdisciplinary collaboration in HAII is critical to enabling ethical, transparent, and effective AI systems that serve human needs.
• Promoting HAII research and application strengthens the foundation for developing trustworthy and human-centered AI systems.