{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_52806"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_52806","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"JUST AND TRUSTWORTHY APPLICATIONS OF DYNAMIC ARTIFICIAL INTELLIGENCE","abstract":"Thesis (Ph.D.)--Michigan State University. Philosophy - Doctor of Philosophy, 2025","abstract_html":"Thesis (Ph.D.)--Michigan State University. Philosophy - Doctor of Philosophy, 2025","abstract_has_math":false,"creators":["LaRosa, Emily"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Elliott, Kevin","Danks, David","O'Rourke, Michael","Kendig, Catherine","Ferkany, Matthew"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-11-18T18:59:28Z","subjects":["Artificial intelligence"],"languages":["eng"],"rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52806","local:LaRosa_grad.msu_0128D_21823"],"render_values":[{"text":"etd:52806","href":null,"code":true},{"text":"local:LaRosa_grad.msu_0128D_21823","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/9msj-7g28","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/9msj-7g28","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_52806","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Elliott, Kevin","Danks, David","O'Rourke, Michael","Kendig, Catherine","Ferkany, Matthew"]},{"key":"dc:creator","label":"Author","values":["LaRosa, Emily"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial intelligence"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial-NoDerivatives 4.0 International"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52806","local:LaRosa_grad.msu_0128D_21823","https://doi.org/doi:10.25335/9msj-7g28"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (Ph.D.)--Michigan State University. Philosophy - Doctor of Philosophy, 2025","The overarching goal of this dissertation is to contribute to more trustworthy governanceof dynamic AI systems. The general question is: what is a trustworthy approach to governance of dynamic, world-altering systems? Dynamic AI systems are those whose behavior continues to evolve post-deployment through retraining, adaptation, and environmental interaction. Unlike technologies such as pharmaceuticals or GMOs, they reshape the very contexts in which they are embedded, making them difficult to regulate through conventional frameworks. These systems pose challenges of trust, responsibility, and transparency that cannot be resolved through technical reliability alone. My dissertation develops normative and epistemic frameworks to meet these challenges, moving from interpersonal dynamics in healthcare AI, to systemic transparency strategies through Curated Information Frameworks (CIFs), and finally to the many governance challenges of Agricultural AI (AgAI), which exemplify the complex entanglements of these systems. Chapter 2, An Algorithm in Doctor’s Clothing, focuses on the role of trust in the deployment of healthcare AI. Moving from Annette Baier’s (1986) account of trust, I argue that current AI systems lack the agency necessary to be trusted in their own right. Instead, what I term extensional, agentially-grounded trust is needed: trust anchored in a responsible human agent (e.g., the physician as an expert), whose moral and professional obligations extend to encompass the AI system. This preserves patient-centered medicine by ensuring that discretionary judgment remains with a human who can respect patient values and autonomy. Chapter 3, Curated Information and Agricultural AI, introduces the Curated Information Framework (CIF) as a new model for appropriate, justifiable transparency. CIFs reject the extremes of “black box” opacity and “white box” full disclosure. Instead, they structure information in a way that is curated, digestible, and actionable. Rather than assuming that more transparency is always beneficial, CIFs provide role-specific, context-sensitive information sharing tailored to what different stakeholders actually need and find useful. Chapter 4, Distinctive Aspects of Agricultural AI, explains why AgAI requires unique governance. Unlike other AI systems, AgAI is materially embedded, hyperlocal, and socially entangled: it interacts directly with land, ecosystems, labor, and community traditions. I propose a taxonomy of AgAI (private, public, and security-facing systems) and show why regulatory analogies from pharmaceuticals or GMOs fail to capture AgAI’s distinctive risks. Governance of AgAI ought to address three features: materiality, context-sensitivity, and entanglement. I argue that governance of AgAI must be materially reflexive, epistemically plural, and participatory, drawing on local expertise and Traditional Ecological Knowledge (TEK) alongside scientific methods. My dissertation demonstrates that trustworthy AI governance requires a layered approach: moral, epistemic, and institutional. First, trust in AI must be grounded in human agents, not the systems themselves (Chapter 2). Second, information about AI systems must be curated to the needs and capacities of diverse stakeholders, avoiding both opacity and overload (Chapter 3). Third, governance frameworks must be tailored to the unique challenges of dynamic, materially embedded and entangled systems such as AgAI, which demand participatory and pluralist oversight (Chapter 4). From these analyses, I conclude that an adequate governance system for dynamic AI such as AgAI ought to pursue five characteristics: hybrid participatory governance; contextual CIFs; post-deployment monitoring; protections for local knowledge; and agroecological and cultural impact assessments (Chapter 5).","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (169 p.) : ill."]},{"key":"dc:title","label":"Title","values":["JUST AND TRUSTWORTHY APPLICATIONS OF DYNAMIC ARTIFICIAL INTELLIGENCE"]}]}],"canonical_facts":{"dc:contributor":["Elliott, Kevin","Danks, David","O'Rourke, Michael","Kendig, Catherine","Ferkany, Matthew"],"dc:creator":["LaRosa, Emily"],"dc:date":["2025"],"dc:description":["Thesis (Ph.D.)--Michigan State University. Philosophy - Doctor of Philosophy, 2025","The overarching goal of this dissertation is to contribute to more trustworthy governanceof dynamic AI systems. The general question is: what is a trustworthy approach to governance of dynamic, world-altering systems? Dynamic AI systems are those whose behavior continues to evolve post-deployment through retraining, adaptation, and environmental interaction. Unlike technologies such as pharmaceuticals or GMOs, they reshape the very contexts in which they are embedded, making them difficult to regulate through conventional frameworks. These systems pose challenges of trust, responsibility, and transparency that cannot be resolved through technical reliability alone. My dissertation develops normative and epistemic frameworks to meet these challenges, moving from interpersonal dynamics in healthcare AI, to systemic transparency strategies through Curated Information Frameworks (CIFs), and finally to the many governance challenges of Agricultural AI (AgAI), which exemplify the complex entanglements of these systems. Chapter 2, An Algorithm in Doctor’s Clothing, focuses on the role of trust in the deployment of healthcare AI. Moving from Annette Baier’s (1986) account of trust, I argue that current AI systems lack the agency necessary to be trusted in their own right. Instead, what I term extensional, agentially-grounded trust is needed: trust anchored in a responsible human agent (e.g., the physician as an expert), whose moral and professional obligations extend to encompass the AI system. This preserves patient-centered medicine by ensuring that discretionary judgment remains with a human who can respect patient values and autonomy. Chapter 3, Curated Information and Agricultural AI, introduces the Curated Information Framework (CIF) as a new model for appropriate, justifiable transparency. CIFs reject the extremes of “black box” opacity and “white box” full disclosure. Instead, they structure information in a way that is curated, digestible, and actionable. Rather than assuming that more transparency is always beneficial, CIFs provide role-specific, context-sensitive information sharing tailored to what different stakeholders actually need and find useful. Chapter 4, Distinctive Aspects of Agricultural AI, explains why AgAI requires unique governance. Unlike other AI systems, AgAI is materially embedded, hyperlocal, and socially entangled: it interacts directly with land, ecosystems, labor, and community traditions. I propose a taxonomy of AgAI (private, public, and security-facing systems) and show why regulatory analogies from pharmaceuticals or GMOs fail to capture AgAI’s distinctive risks. Governance of AgAI ought to address three features: materiality, context-sensitivity, and entanglement. I argue that governance of AgAI must be materially reflexive, epistemically plural, and participatory, drawing on local expertise and Traditional Ecological Knowledge (TEK) alongside scientific methods. My dissertation demonstrates that trustworthy AI governance requires a layered approach: moral, epistemic, and institutional. First, trust in AI must be grounded in human agents, not the systems themselves (Chapter 2). Second, information about AI systems must be curated to the needs and capacities of diverse stakeholders, avoiding both opacity and overload (Chapter 3). Third, governance frameworks must be tailored to the unique challenges of dynamic, materially embedded and entangled systems such as AgAI, which demand participatory and pluralist oversight (Chapter 4). From these analyses, I conclude that an adequate governance system for dynamic AI such as AgAI ought to pursue five characteristics: hybrid participatory governance; contextual CIFs; post-deployment monitoring; protections for local knowledge; and agroecological and cultural impact assessments (Chapter 5).","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (169 p.) : ill."],"dc:identifier":["etd:52806","local:LaRosa_grad.msu_0128D_21823","https://doi.org/doi:10.25335/9msj-7g28"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"dc:subject":["Artificial intelligence"],"dc:title":["JUST AND TRUSTWORTHY APPLICATIONS OF DYNAMIC ARTIFICIAL INTELLIGENCE"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:28Z"}