[Paper Review] Proceedings of The first international workshop on eXplainable AI for the Arts (XAIxArts)

Nick Bryan–Kinns, Corey Ford+7 more|arXiv (Cornell University)|Oct 10, 2023

Explainable Artificial Intelligence (XAI)Computer Science3 citations

TL;DR

This workshop proceedings present the foundational discourse on explainable AI (XAI) in the arts, bringing together HCI, AI, and digital arts researchers to explore how XAI can enhance transparency and interpretability in artistic AI systems. The key contribution is establishing XAIxArts as a new interdisciplinary forum for advancing ethical, human-centered AI in creative practices.

ABSTRACT

This first international workshop on explainable AI for the Arts (XAIxArts) brought together a community of researchers in HCI, Interaction Design, AI, explainable AI (XAI), and digital arts to explore the role of XAI for the Arts. Workshop held at the 15th ACM Conference on Creativity and Cognition (C&C 2023).

Motivation & Objective

• To establish a dedicated interdisciplinary forum for researchers working at the intersection of explainable AI (XAI) and the arts.
• To address the growing need for transparency and interpretability in AI-driven artistic processes and outputs.
• To explore how XAI can support artists, designers, and audiences in understanding and engaging with AI-generated or AI-augmented art.
• To identify key challenges, ethical considerations, and design principles for XAI in creative domains.
• To foster collaboration across HCI, interaction design, AI, and digital arts to shape future research and applications in XAI for the arts.

Proposed method

• Organized a full-day international workshop at the 15th ACM Conference on Creativity and Cognition (C&C 2023) to bring together experts from diverse fields.
• Facilitated panel discussions, lightning talks, and collaborative sessions focused on XAI in artistic contexts.
• Collected insights and emerging themes through structured brainstorming and group synthesis activities.
• Solicited contributions from researchers in HCI, interaction design, AI, XAI, and digital arts to ensure interdisciplinary breadth.
• Documented key findings and research directions through a curated proceedings publication.
• Established a foundation for future research by identifying core themes, challenges, and opportunities in XAI for the arts.

Experimental results

Research questions

• RQ1 How can explainable AI enhance transparency and user understanding in AI-augmented artistic practices?
• RQ2 What are the unique interpretability and ethical challenges when applying XAI to creative domains like music, visual arts, and performance?
• RQ3 How can XAI tools be co-designed with artists and audiences to support meaningful engagement with AI-generated art?
• RQ4 What design principles and interaction patterns are most effective for explaining AI processes in artistic contexts?
• RQ5 How can interdisciplinary collaboration between XAI and arts researchers lead to more responsible and inclusive AI in creative applications?

Key findings

• The workshop successfully established XAIxArts as a new interdisciplinary forum for advancing XAI in artistic and creative contexts.
• There is a strong consensus that explainability in AI for the arts must go beyond technical transparency to include aesthetic, emotional, and contextual understanding.
• Artists and audiences require explanations that are interpretable, context-sensitive, and aligned with creative intent, not just model accuracy.
• Key challenges include balancing technical explainability with artistic ambiguity, and ensuring that explanations do not oversimplify complex creative processes.
• The workshop identified a need for new evaluation frameworks that consider both interpretability and artistic value in AI-generated works.
• Collaborative, participatory design approaches were emphasized as essential for developing XAI tools that serve diverse artistic communities effectively.