Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor
By Hannah Lebovits. 2018
2,275 citations recorded by the source corpus.
Almost two decades into the new millennium, it is unlikely that the use of digital technologies will slow in any significant way, particularly in the public sector. As local and regional public age...
Subfields: Sociology and Political Science, Social Sciences & Humanities
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Sociology and Political Science
- Maranke Wieringa
- Joshua A. Kroll
- Emilio Ferrara
- Emily L. Denton
- M. Fourcade
- Sophie Bishop
- A. Hanna
- Vincent C. Müller
In Social Sciences & Humanities
- Ben Green
- Moritz Hardt
- Nitin Kohli
- Emilio Ferrara
- Joshua A. Kroll
- Jenna Burrell
- D. boyd
- Solon Barocas
Citations in the corpus
- Diagnosing Gender Bias in Image Recognition Systems (2020, methodology): “Although prior work has sought to either use algorithms (Anastasopoulos et al. 2016; Casas and Webb Williams 2018; Xi et al. 2019) or identify biases in them (Buolamwini and Gebru 2018; Crawford and Paglen 2019; Eubanks 2018), we argue that it is critical for scholars to do both at the same time.”
- Fairness and Abstraction in Sociotechnical Systems (2019, background): “…relevant social groups can be obviated by taking seriously the needs of people typically un-derrepresented in these processes, rather than making assumptions about their welfare [23], and by understanding the power dynamics that prevent these voices from having influence in society to begin with.”
- Large image datasets: A pyrrhic win for computer vision? (2020, background): “Image classification and labelling practices, without the necessary precautions and awareness of these problematic histories, pick up these stereotypes and prejudices and perpetuate them [31, 58, 59].”
- Bringing the People Back In: Contesting Benchmark Machine Learning Datasets (2020, background): “…systems abound in ways that they have failed people of color (Noble, 2018; Benjamin, 2019), women (Bolukbasi et al., 2016), LGBTQ+ communities (Scheuerman et al., 2019), people with disabilities (Hutchinson et al., 2020; Trewin, 2018), and the working class and those in poverty (Eubanks, 2018).”
- Garbage in, garbage out?: do machine learning application papers in social computing report where human-labeled training data comes from? (2019, neutral): “This is even the case with relatively simple forms of analytics and automation — such as if-then statements, linear regressions, or rule-based expert systems [11, 62].”
- Participatory Problem Formulation for Fairer Machine Learning Through Community Based System Dynamics (2020, background): “…ML practitioners — often lack the lived experiences required to comprehensively approximate and account for the various peripheral stakeholders their interventions will impact, especially the communities that are the most vulnerable to unfair outcomes (Eubanks, 2018; Campolo et al., 2017).”
- The Disparate Effects of Strategic Manipulation (2018, background): “Theoretically, algorithms have great potential to reform existing systems to become both more efficient and equitable, but as exposed by various high-profile investigations [1, 2, 3, 4], prediction-based models that are deployed to make or assist with consequential social decisions are, in practice,”
- The Society of Algorithms (2021, background): “The automation of social services eligibility criminalizes the most vulnerable families and diverts them from the benefits to which they are entitled (Eubanks 2017).”
- Managing extreme AI risks amid rapid progress (2023, methodology): “Inclusive AI development: AI advancement will need methods to mitigate biases and integrate the values of the many populations it will affect 20,55 .”
- ORES (2019, neutral): “Algorithmic systems [94] play increasingly crucial roles in the governance of social processes [7, 26, 40].”
- Truth from the machine: artificial intelligence and the materialization of identity (2021, background): “Both theorists and practitioners have demonstrated ways in which algorithmic systems produce disparate impacts for di ff erent populations – impacts which frequently negatively e ff ect trans people, people of colour, and/or the poor (Eubanks 2018; Noble 2018; Keyes 2018).”
- When data is capital: Datafication, accumulation, and extraction (2019, background): “At times, data is disproportionally extracted from certain groups, such as when poor people of colour are subjected to systematic tracking by government agencies and financial institutions (Eubanks, 2018).”
- Managing visibility on YouTube through algorithmic gossip (2019, background): “Although Victory cannot prove discrimination, algorithms do often sustain bias towards marginalised people (Eubanks, 2017).”
- Algorithmic realism: expanding the boundaries of algorithmic thought (2020, background): “A great deal of other recent work has documented the particular ways in which the design, application, and use of algorithms can exacerbate marginalization and inequality [18, 42, 53, 64, 107, 110].”