The Role of Interactive Visualization in Fostering Trust in AI.

Area of Science:

• Computer Science
• Human-Computer Interaction
• Artificial Intelligence Ethics

Background:

• Growing adoption of AI necessitates focus on AI trustworthiness, fairness, interpretability, and accountability.
• Building public trust in AI systems is a critical societal challenge.
• Existing approaches to AI trust often overlook the role of user interaction.

Purpose of the Study:

• To explore the role of interactive visualization in fostering trust in artificial intelligence (AI).
• To present a manifesto on the relevance and limitations of visualization in AI trust.
• To assert that trust in AI is multifaceted and requires human-centered approaches.

Main Methods:

• Conceptual analysis and manifesto development.
• Discussion of AI trustworthiness dimensions.
• Examination of interactive visualization's potential impact on AI trust.

Main Results:

• Trust in AI is not solely a technical issue but a dynamic, evolving human perception.
• Interactive visualization is crucial for enabling human agency within AI systems.
• Visualization can address certain aspects of AI trust but cannot solve all challenges.

Conclusions:

• Interactive visualization is essential for enhancing human understanding and control over AI.
• Addressing AI trust requires a holistic approach that includes visualization alongside technical and ethical considerations.
• Future research should focus on developing and evaluating interactive visualization techniques for trustworthy AI.