본문

서브메뉴

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 Go...
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
저자명  
Whitney, Cedric Deslandes.
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  
Machine unlearning
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


    New Books MORE
    Statistics for the past 3 years. Go to brief

    Подробнее информация.

    • Бронирование
    • не существует
    • моя папка
    • Первый запрос зрения
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    материал
    Reg No. Количество платежных Местоположение статус Ленд информации
    TF19305 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * Бронирование доступны в заимствований книги. Чтобы сделать предварительный заказ, пожалуйста, нажмите кнопку бронирование

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.