Ethotic Heuristics in Artificial Intelligence: A Rhetorical Framework for Guiding Responsible Data Design Praxis in Healthcare and Surveillance

Abstract

This thesis investigates the convergence of artificial intelligence (AI), human-centered design, and rhetoric across three interconnected essays. This dissertation centers on design heuristics as its primary analytic and unifying framework, drawing from traditions such as Data Feminism and rhetorical inquiry. It explores three interrelated domains: (1) AI-driven human-computer interaction (HCI) design; (2) the implications of AI-powered design for women’s health privacy, particularly in the post-Roe v. Wade U.S. context; and (3) critical discourse surrounding AI in surveillance technologies. Using a multi-method approach—including rhetorical analysis, Critical Discourse Analysis (CDA), case studies, and stakeholder perspectives—this research interrogates how AI systems construct algorithmic ethopoeic representations that commodify user data. The first essay introduces a set of practical heuristics for HCI designers by integrating principles from Design Thinking, thereby fostering ethical dialogue, and strengthening human-centered approaches in the context of rapid AI development. The second essay employs rhetorical analysis to examine the construction of “algorithmic ethopoeia”—the process through which AI systems perform moral characterization through data practices, design choices, and institutional logics—within sensitive socio-technical domains. Algorithmic ethopoeia is a concept central to this dissertation and is defined in more detail on page 3 of this essay. By foregrounding this concept, the essay emphasizes the urgent need for robust protections surrounding personal data integrity and highlights how algorithmic systems actively participate in shaping judgments about identity, risk, and responsibility. This section, grounded in Data Feminism, empowers designers, activists, and policymakers to advocate for more secure and transparent AI applications, particularly in the domain of women’s health privacy. The final essay employs CDA to critique the discourse surrounding AI-driven surveillance, focusing on predictive policing and facial recognition technologies. Through the analysis of competing narratives and stakeholder perspectives, it reveals ethical dilemmas related to systemic biases and authoritarian practices, arguing for rigorous oversight and regulatory frameworks. Surveillance contextual heuristics are proposed to guide the responsible deployment of AI in public safety while safeguarding civil liberties. Collectively, these investigations underscore the imperative for ethical, context-sensitive, and rigorously informed design heuristics to guide the responsible integration of AI across diverse domains. They advance the discourse on user privacy, regulatory compliance, and human-centered innovation, while simultaneously promoting the development of design practices that are both ethically sound and equitable.