AI-Generated Images for representing Individuals: Navigating the Thin Line Between Care and Bias

2025-09-03

AI Summary

This study examines the ethical tensions arising from AI-generated human portraits in data storytelling—specifically, how to humanize depression statistics while upholding empathy and avoiding stereotypical bias. We developed a public communication prototype for European depression data by integrating interactive data visualization with generative AI portrait synthesis, and conducted user experiments to systematically assess audience emotional responses and ethical concerns regarding these AI portraits. Our key contribution is the first conceptualization of the ethical boundaries of figurative representation at the intersection of data visualization and generative AI, proposing a “reflexive figuration” framework. This framework provides both theoretical guidance and practical design principles for responsible visual representation in science communication, advancing data storytelling from technical feasibility toward ethical intentionality. (136 words)

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

This research discusses the figurative tensions that arise when using portraits to represent individuals behind a dataset. In the broader effort to communicate European data related to depression, the Kiel Science Communication Network (KielSCN) team attempted to engage a wider audience by combining interactive data graphics with AI-generated images of people. This article examines the project's decisions and results, reflecting on the reaction from the audience when information design incorporates figurative representations of individuals within the data.

Problem

• Addressing bias in AI-generated portrait representations
• Balancing care and bias in data visualization
• Examining audience reactions to figurative data representations

Innovation

• AI-generated images combined with interactive data graphics
• Representing individuals behind datasets using portraits
• Navigating bias and care in figurative representations