서브메뉴
검색
Algorithms as Forward Salients: Examining the Politics of Algorithmic Infrastructure
Algorithms as Forward Salients: Examining the Politics of Algorithmic Infrastructure
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105232
- ISBN
- 9798291567326
- DDC
- 020
- 저자명
- Grill, Gabriel.
- 서명/저자
- Algorithms as Forward Salients: Examining the Politics of Algorithmic Infrastructure
- 발행사항
- [Sl] : University of Michigan, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 208 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-02, Section: B.
- 주기사항
- Advisor: Lindtner, Silvia;Sandvig, Christian.
- 학위논문주기
- Thesis (Ph.D.)--University of Michigan, 2025.
- 초록/해제
- 요약As algorithmic technologies have increasingly been adopted in industry and the public sector, they have also become more essential to how we live, work, and govern. This dissertation examines what it means for algorithms to become increasingly infrastructural, and, conversely, for contemporary infrastructures to become more algorithmic. This work contributes to infrastructure and algorithm studies, as well as to the fields of science and technology and information studies, by offering the following theoretical and empirical insights. First, I introduce the analytic forward salient to describe how algorithms are promoted as the next frontier, thus marking established infrastructures and practices as problems in need of intervention. Second, I contribute the conceptual hybrid of algorithmic infrastructure by unpacking its characteristics and the contradictions that emerge as algorithms and infrastructures are intertwined. Finally, I apply these concepts to analyze three case studies that focus on algorithmic technologies intended to alter and supplant existing infrastructure. I unpack the social and political consequences of their adoption, such as the undermining of public interest infrastructure, and how it is justified. The first case concerns an algorithmic profiling tool intended to assist caseworkers of a public employment agency in making decisions about the allocation of support measures during unemployment. The algorithm became a controversy as researchers and activists highlighted how it can reinforce inequalities along dimensions such as gender, class, and disability. My analysis unpacks the various biases of the system and reflects on how these identified biases also acted as boundary objects that enabled disparate groups, like activists, researchers, and legal experts, to challenge the system. The second case examines research and software products that promise to detect and predict social and political unrest by analyzing social media data to aid risk analysts in mitigating the impact of disruptions. I highlight how these systems expand algorithmic surveillance, point to underlying faulty assumptions around the power of big data, and discuss how they construct risk in ways that produce increased suspicion about grievances voiced online. The third case centers on globalized testing infrastructures intended to enhance generative AI products and produce evaluation results that certify their capabilities. I unpack how technology companies benefit from incentive structures and platforms that promote and mask under- and unpaid testing labor and point to structural issues in current evaluation and testing practices that create overly optimistic impressions of the capabilities of AI systems, fueling hype and misguided adoption. Taken together, this dissertation provides insights into a concerning decline of "infrastructural values" such as the subversion of the public interest orientation of many infrastructures. The rise of algorithmic technologies as forward salients promises automation, efficiency, and centralized control, while concentrating power, which can undermine support structures and thereby exacerbate inequalities.
- 일반주제명
- Information science
- 일반주제명
- Philosophy of science
- 일반주제명
- Sociology
- 일반주제명
- Information technology
- 키워드
- Infrastructures
- 키워드
- Algorithms
- 기타저자
- University of Michigan Information
- 기본자료저록
- Dissertations Abstracts International. 87-02B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359893
■00520260202105232
■006m o d
■007cr#unu||||||||
■020 ▼a9798291567326
■035 ▼a(MiAaPQ)AAI32271907
■035 ▼a(MiAaPQ)umichrackham006302
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a020
■1001 ▼aGrill, Gabriel.
■24510▼aAlgorithms as Forward Salients: Examining the Politics of Algorithmic Infrastructure
■260 ▼a[Sl]▼bUniversity of Michigan▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a208 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-02, Section: B.
■500 ▼aAdvisor: Lindtner, Silvia;Sandvig, Christian.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2025.
■520 ▼aAs algorithmic technologies have increasingly been adopted in industry and the public sector, they have also become more essential to how we live, work, and govern. This dissertation examines what it means for algorithms to become increasingly infrastructural, and, conversely, for contemporary infrastructures to become more algorithmic. This work contributes to infrastructure and algorithm studies, as well as to the fields of science and technology and information studies, by offering the following theoretical and empirical insights. First, I introduce the analytic forward salient to describe how algorithms are promoted as the next frontier, thus marking established infrastructures and practices as problems in need of intervention. Second, I contribute the conceptual hybrid of algorithmic infrastructure by unpacking its characteristics and the contradictions that emerge as algorithms and infrastructures are intertwined. Finally, I apply these concepts to analyze three case studies that focus on algorithmic technologies intended to alter and supplant existing infrastructure. I unpack the social and political consequences of their adoption, such as the undermining of public interest infrastructure, and how it is justified. The first case concerns an algorithmic profiling tool intended to assist caseworkers of a public employment agency in making decisions about the allocation of support measures during unemployment. The algorithm became a controversy as researchers and activists highlighted how it can reinforce inequalities along dimensions such as gender, class, and disability. My analysis unpacks the various biases of the system and reflects on how these identified biases also acted as boundary objects that enabled disparate groups, like activists, researchers, and legal experts, to challenge the system. The second case examines research and software products that promise to detect and predict social and political unrest by analyzing social media data to aid risk analysts in mitigating the impact of disruptions. I highlight how these systems expand algorithmic surveillance, point to underlying faulty assumptions around the power of big data, and discuss how they construct risk in ways that produce increased suspicion about grievances voiced online. The third case centers on globalized testing infrastructures intended to enhance generative AI products and produce evaluation results that certify their capabilities. I unpack how technology companies benefit from incentive structures and platforms that promote and mask under- and unpaid testing labor and point to structural issues in current evaluation and testing practices that create overly optimistic impressions of the capabilities of AI systems, fueling hype and misguided adoption. Taken together, this dissertation provides insights into a concerning decline of "infrastructural values" such as the subversion of the public interest orientation of many infrastructures. The rise of algorithmic technologies as forward salients promises automation, efficiency, and centralized control, while concentrating power, which can undermine support structures and thereby exacerbate inequalities.
■590 ▼aSchool code: 0127.
■650 4▼aInformation science
■650 4▼aPhilosophy of science
■650 4▼aSociology
■650 4▼aInformation technology
■653 ▼aInfrastructures
■653 ▼aAlgorithms
■653 ▼aScience and technology studies
■653 ▼aTechnology companies
■653 ▼aInformation studies
■690 ▼a0723
■690 ▼a0402
■690 ▼a0489
■690 ▼a0626
■71020▼aUniversity of Michigan▼bInformation.
■7730 ▼tDissertations Abstracts International▼g87-02B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359893▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


