서브메뉴
검색
From Data to Evaluation: Investigating the Limits of the Individual-Control Model in AI Governance
From Data to Evaluation: Investigating the Limits of the Individual-Control Model in AI Governance
Detailed Information
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202103555
- ISBN
- 9798288862052
- DDC
- 020
- 서명/저자
- From Data to Evaluation: Investigating the Limits of the Individual-Control Model in AI Governance
- 발행사항
- [Sl] : University of California, Berkeley, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 143 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-01, Section: B.
- 주기사항
- Advisor: Mulligan, Deirdre.
- 학위논문주기
- Thesis (Ph.D.)--University of California, Berkeley, 2025.
- 초록/해제
- 요약Technical fixes for AI harm are usually framed through an individual‑control lens: give people notice, let them opt out, erase their row, and the problem is solved. Building on relational theories of data governance, this dissertation probes the limits of that premise across the full machine‑learning pipeline. It traces three tightly coupled 'critical cases' that sit at successive system layers. At the input level, a forensic audit of privacy‑branded synthetic‑data workflows shows how nominal anonymity can still magnify racial mis‑representation. At the model level, a scenario‑based study of machine unlearning reveals that participants' wish to be erased is conditioned by collective‑risk calculations, not by autonomy alone. At the evaluation level, a meta‑analysis of contextual benchmarks demonstrates that large language models which pass instruction‑following tests routinely fail when placed in norm‑conflict or long‑context settings.Across these settings the findings converge on four recurring failure modes. First, individual row‑level privacy gains often intensify group harms, exposing relational blind‑spots. Second, generative and unlearning tools rupture data lineage, undermining regulatory oversight that depends on provenance. Third, performance that looks robust in context‑free scorecards collapses under realistic framing, revealing deep contextual brittleness. Finally, the operational cost of verifying compliance with individual‑control mechanisms creates enforcement frictions that outstrip regulatory capacity.Methodologically, the work illustrates how combining dataset forensics, interpretive user research, and quantitative stress‑testing produces a multi‑level socio‑technical diagnosis unavailable to any single approach. In doing so it heeds recent calls for a richer 'evaluation science' in AI ethics and charts a research agenda toward lineage‑aware tooling, context‑sensitive benchmarks, and governance structures capable of addressing the relational nature of machine‑learning harm.
- 일반주제명
- Information science
- 일반주제명
- Computer science
- 일반주제명
- Information technology
- 키워드
- Context
- 키워드
- Evaluation
- 키워드
- Machine learning
- 키워드
- Model deletion
- 키워드
- Synthetic data
- 기타저자
- University of California, Berkeley Information Management & Systems
- 기본자료저록
- Dissertations Abstracts International. 87-01B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017357750
■00520260202103555
■006m o d
■007cr#unu||||||||
■020 ▼a9798288862052
■035 ▼a(MiAaPQ)AAI32042031
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a020
■1001 ▼aWhitney, Cedric Deslandes.
■24510▼aFrom Data to Evaluation: Investigating the Limits of the Individual-Control Model in AI Governance
■260 ▼a[Sl]▼bUniversity of California, Berkeley▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a143 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-01, Section: B.
■500 ▼aAdvisor: Mulligan, Deirdre.
■5021 ▼aThesis (Ph.D.)--University of California, Berkeley, 2025.
■520 ▼aTechnical fixes for AI harm are usually framed through an individual‑control lens: give people notice, let them opt out, erase their row, and the problem is solved. Building on relational theories of data governance, this dissertation probes the limits of that premise across the full machine‑learning pipeline. It traces three tightly coupled 'critical cases' that sit at successive system layers. At the input level, a forensic audit of privacy‑branded synthetic‑data workflows shows how nominal anonymity can still magnify racial mis‑representation. At the model level, a scenario‑based study of machine unlearning reveals that participants' wish to be erased is conditioned by collective‑risk calculations, not by autonomy alone. At the evaluation level, a meta‑analysis of contextual benchmarks demonstrates that large language models which pass instruction‑following tests routinely fail when placed in norm‑conflict or long‑context settings.Across these settings the findings converge on four recurring failure modes. First, individual row‑level privacy gains often intensify group harms, exposing relational blind‑spots. Second, generative and unlearning tools rupture data lineage, undermining regulatory oversight that depends on provenance. Third, performance that looks robust in context‑free scorecards collapses under realistic framing, revealing deep contextual brittleness. Finally, the operational cost of verifying compliance with individual‑control mechanisms creates enforcement frictions that outstrip regulatory capacity.Methodologically, the work illustrates how combining dataset forensics, interpretive user research, and quantitative stress‑testing produces a multi‑level socio‑technical diagnosis unavailable to any single approach. In doing so it heeds recent calls for a richer 'evaluation science' in AI ethics and charts a research agenda toward lineage‑aware tooling, context‑sensitive benchmarks, and governance structures capable of addressing the relational nature of machine‑learning harm.
■590 ▼aSchool code: 0028.
■650 4▼aInformation science
■650 4▼aComputer science
■650 4▼aInformation technology
■653 ▼aContext
■653 ▼aEvaluation
■653 ▼aMachine learning
■653 ▼aMachine unlearning
■653 ▼aModel deletion
■653 ▼aSynthetic data
■690 ▼a0723
■690 ▼a0984
■690 ▼a0489
■690 ▼a0800
■71020▼aUniversity of California, Berkeley▼bInformation Management & Systems.
■7730 ▼tDissertations Abstracts International▼g87-01B.
■790 ▼a0028
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17357750▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
Preview
Export
ChatGPT Discussion
AI Recommended Related Books
Подробнее информация.
- Бронирование
- не существует
- моя папка
- Первый запрос зрения
- Non-Book Loan Application
- Nighttime Book Loan Application
Available after logging in.


