Automating Inequality: How High-Tech Tools Profile, Police, and Punish The Poor

edit

Automating Inequality
AuthorVirginia Eubanks
LanguageEnglish
GenreSocial justice
PublisherMacmillan Publishers
Publication date
Jan 23, 2018
Publication placeUnited States
Media typeHardback and paperback
ISBN9781250074317


Automating Inequality: How High-Tech Tools Profile, Police, and Punish The Poor, is a 2018 non-fiction book written by Virginia Eubanks, an associate professor of Political Science at the University of Albany. The book which has been described as “riveting” by Liza Featherstone of the New York Times [1] delves into how governmental institutions use technology, especially Artificial Intelligence for automated algorithmic decision-making. The focal point of the book discusses how the system frames and treats the poor as "unreliable, untrustworthy and worthless" [2] and how the end results of these algorithmic decisions further marginalize the poor and those on the brink of poverty [3].

Introduction

edit

The introductory chapter of the book titled "Red flags" discusses Eubanks’ experience with an insurance company after her partner was assaulted and needed long term medical care. Eubanks believed that the algorithm used by their insurance provider red flagged her account, which led her to suffer serious emotional stress, debt, and a lot of back and forth with the provider before it was eventually resolved [4].

The use of artificial intelligence in different sectors are " frequently presented as some of the most promising, and potentially impactful, technological advances available to humankind today " [5]. The benefits supposedly include eradicating subjectivity and bias [6], in addition to saving time and promoting efficiency. However, as Eubanks explained, algorithmic decision-making is not quite the solution it’s been touted to be and has actually been making things worse. Its major drawbacks include, discrimination, bias and fostering inequality [7]. Therefore, as pointed out by Julia Manero, "when technologies reinforces inequality, they are no longer neutral” [8] and should not be tagged as objective. In addition, the decision-making algorithm replicates the biases, systems, and stereotypical beliefs that are already in place [9] because algorithms do not exist in a vacuum; they are created by humans, who decides what information is used as criteria for the decision-making process. Hence, if an algorithm perpetuates racism, it is because the system was racist to begin with [10].

The result of the investigation carried out by Eubanks revealed the trend of how the poor have been treated over the years; like criminals on trial, surveyed, investigated by caseworkers and subjected to policies that exacerbate their poverty. The automated algorithmic decision adopted by the city of Los Angeles also followed this trend. Therefore, rather than helping, further intensified discrimination, and encouraged the profiling and surveillance of the poor and homeless[4].

Another major theme explored in the introductory chapter is the relationship between technology and poverty. Prior to surveillance technology becoming ubiquitous, the government tracked the poor with assistance schemes. The EBT card for instance is routinely used to track the movement and purchase history of its users [4]. As predicted in the book, that eventually the tracking system designed for the poor will be used on everyone, surveillance tools are now deeply embedded in the society; used by the government, tech giants like Amazon (company) and Google, academic institutions and even retailers.

Eubanks also pointed out that while surveillance technology might be ubiquitous, the poor and marginalized groups are mostly affected and suffer the consequences of being targeted, investigated, and discriminated against [4]. The result of this surveillance and automated decisions are made palatable for the government and institutions who they are able to hide behind a faceless algorithm when making inhuman choices [7].

High-Tech Homelessness in the City of Angels

edit

The third chapter of Eubanks’ book is titled "High Tech Homelessness in the City of Angels" describing the history and the struggles of inhabitants of the area known as Skid Row in Los Angeles. According to Eubanks, Skid Row, used to be a thriving neighborhood inhabited by people of color and poor whites till most of it was demolished in order to make room for more homes to be built to cater to the growing population. However, when it was time to replace what had been demolished, white middle- class Angelenos kicked against the idea, calling the plan to increase low-income housing in Los Angeles, “a Red plot to control LA housing” [4]. Therefore, LA built only a small fraction of public housing of other cities its size.

In addition, in the 1960s the number of affordable housing units in Skid Row was reduced from 15000 units to about 7500 units. A similar plan was proposed in the 1970s called the "Silver Book plan" which suggested that what was left of the Skid Row be demolished to make room for extensions of the University of Southern Carolina and the University of Carolina. Residents of the Skid Row were to be sent to a massive detoxification and rehabilitation center. This plan was kicked against by activists and residents who proposed an alternative, the "Blue Book plan" which protected the remaining single room occupancy hotels on Skid Row and encouraged the government and local nonprofit to commit resources to improving housing and social services in the area. As highlighted by Eubanks, the Blue Book plan prevailed because Skid Row was presented as lawless and frightening, and this was pivotal to the arguments of activists who threatened that a wave of homeless and indigent people would be unleashed on the suburban neighborhoods of Los Angeles if Skid Row was demolished [4].

The "Blue Book plan" worked for about four decades. However, due to urbanization and the luxury rental building boom, parts of Skid Row were affected over the years, “…about 16 square blocks- a third of its size -in ten years” [4]. The Skid Row today is a juxtaposition of rich and poor with the rich living in expensive buildings while the poor live in makeshift tents. The ability for the rich and poor to co exist have been commended. However it is obvious this might be temporary, as Eubanks asserted, for instance, efforts to turn the long-abandoned Cecil Hotel into a housing complex for chronically homeless people were stopped by county supervisors in 2014 [4].

Eubank deduced from conversations with the residents and her study of Skid Row that although the people of Skid Row might reside in tents or have less than adequate living situations, they were decent people and not all of them were mentally unstable and dangerous as their neighbors, and the government tried to portray them to be [4].

The Los Angeles Coordinated Entry System

edit

The Los Angeles Coordinated Entry System has been described as the match.com of homeless services. The system is designed to match the unhoused with appropriate available housing and is supported by powerful entities like the United States Department of Housing and Urban Development, Bill & Melinda Gates Foundation and some other homeless service providers. Prior to instituting the Coordinated Entry System, “unhoused people navigated a complex system of waitlists and social service program requiring a great deal of patience, fortitude, and luck” [4]. Coordinated Entry is based on two philosophies: "Prioritization and Housing First.

Prioritization is based on the research of Dennis Culhane of the University of Pennsylvania [11]. His research differentiates between the two types of Homelessness: "Chronic and Crisis". Crisis homelessness indicate that the situation could be temporary with the right measures taken as the individual is usually experiencing a short-term emergency such as job loss, eviction and would often self correct after a short stay at a shelter. Chronic homelessness on the other hand means that the individual tends to be homeless for a long stretch of time, have a disability and/or health problems and have complex support systems. This category of people require permanent housing and other resources [4].

Housing First focused on the providing affordable housing to individuals and families as fast as possible and then offering them voluntary support as opposed to the housing readiness that was operable before which moved the unhoused through different programs and steps like sobriety, treatments, employments before they could be properly housed.

Home For Good

edit

Home for Good is a coordinated entry program launched in 2013 by the United Way of Greater Los Angeles and the Los Angeles Area Chamber of Commerce. The initiative pledged to house 100 of the most vulnerable homeless people on Skid Row in 100 days by combining Prioritization, Housing First, and Technology Forward approaches. In order to accomplish this goal, the VISPDAT(Vulnerability Index -Service Prioritization Decision Assistance Tool) was used to gather extensive personal and intimate information including medical and sexual history, Social Security Number, Immigration status and family situation of the unhoused. Photographs were also taken if the participants consented [4].

The consent forms that were distributed indicated that the information collected will be shared with certain organizations. Request for complete information about the privacy policy would reveal that the information collected will be shared with institutions such as the government, hospitals, LAPD, religious organizations, and the University of California and that the consent was valid for seven years [4].

After the collection of information, the data is stored into the federally approved Homeless Management Information System (HMIS) for the Los Angeles area. The data entered into the HMIS is analyzed by a ranking algorithm and comes up with a score ranging from 1 to 17. Eubanks elucidate that: A “1” would mean that "the person surveyed is low risk and has a relatively small chance of dying or ending up in an emergency room or mental hospital ...and a “17” means the person surveyed is among the most vulnerable" [4].

After a housing vacancy form has been filled out, a second algorithm is run to identify, through the VI-SPDAT score a person who is in greatest need for the available housing and who meets all the stipulated criteria. If a match is made, the eligibility documentation which includes personal and income information is gathered by the housing navigator assigned to the unhoused, then the unhoused fills out an application form with the Housing Authority of the city of Los Angeles(HACLA) who then conducts an interview, verifies the documents, then approves or denies the application. If approved, an apartment is provided to the unhoused, if not, the algorithm is run again to provide a new applicant [4].

Case Studies

edit

Monique Talley and Gary Boatwright are two homeless individuals who took the VI-SPDAT survey. Monique was selected by the algorithm and got placed into a low-cost housing while Gary, even though he took the survey three times, and was living in a tent on the street was not selected by the algorithm. Conversations with both applicants showed that Monique was desperate and ready to talk to anyone about getting housed. She stated “…If it was to get me a roof over my head, I will talk to you, and tell you the truth, and tell you what you want to hear” [4]. Her statement “…(I) will tell you what you want to hear” could be translated to mean that she had an understanding of how the system works and was able come up with the right answers in order to be ranked higher by the algorithm.

Gary on the other hand, was not ready to “bow down” to the system and give up his “self-determination” [4] for a house. From his conversation with Eubanks, it could be inferred that Gary scored lower based on his age and medical history. He confirmed that he was fairly healthy, and Eubanks noticed that he does seem to have a major issue with substance abuse, and does not appear suicidal.

Gary and Monique’s experience with the VI-SPDAT algorithm reinforces the idea that trends in poverty management include public administration setting certain criteria to determine the merits of the poor and singling out those who "deserve" to be helped [12].

Skid Row VS South LA

edit

The South Los Angeles area, though characterized by a significant level of homelessness, received less attention and resources than downtown Los Angeles where Skid Row is located.

The coordinated entry system was seen as a positive development based on experiences of those that were placed into houses successfully like Monique Talley. The limitations of the system became evident in south LA where the houses were simply not available due to limited resource allocation.

The unhoused population of South LA had a different experience with the coordinated entry system. While it’s been established that the algorithm might rank an unhoused lower and therefore deny them access to housing. In south LA, the houses available were significantly lower than those available at Skid Row. Hence, even if a applicant was ranked higher, and aced the HACLA interview, the chances of getting housed within 90 days were very slim [13]

The Comprehensive Homeless Strategy of Mayor Eric Garcetti

edit

The former mayor of Los Angeles, Eric Garcetti proposed a comprehensive homeless strategy released in January 2016. The strategy provides significant support for coordinated entry and promotes rapid re-housing programs for those on the edge of homelessness, offering small amounts of money for expenses like rental assistance, moving costs, and case management. It also supports converting existing commercial structures into short-term bridge housing and provides incentives to encourage landlords to accept Section 8 housing vouchers [13].

Rapid re-housing is based on the Housing First approach, which prioritizes providing individuals and families with immediate access to housing without preconditions such as sobriety or participation in treatment programs. This approach recognizes that stable housing is a foundational element for addressing other challenges faced by homeless individuals and families, such as employment, health, and well-being [13].

Regarding the coordinated entry system in Los Angeles, rapid re-housing is one of the intervention strategies used to match homeless individuals with short-term support, particularly for those who do not qualify for permanent supportive housing but still require assistance to exit homelessness [13].

The Surveillance and Criminalization of the Poor

edit

Data collection, sharing, and surveillance in homeless service programs contribute to the criminalization of the poor and unhoused. The collection of sensitive and personal information, such as social security numbers, medical and domestic violence histories are used to surveil and criminalize the unhoused population particularly because the system equates being poor and homeless to being a criminal [8].

The coordinated entry system in Los Angeles collects and shares personal information with a wide range of organizations, including the Los Angeles Police Department (LAPD), whose officers are able to access these information by simply requesting verbally without a warrant [13]. This raises concerns about the potential misuse of personal information and abuse of office by law enforcement agencies.

The criminalization of homelessness is also worsened by the fact that many basic conditions of homelessness; such as sleeping in public parks or leaving possessions on the sidewalk are officially considered crimes. This can lead to the issuance of tickets, warrants, and arrest of the unhoused population who are usually unable to make bail and end up spending months in jail before the case is decided/dismissed by the court [14]

Old VS New Model of Surveillance

edit

The old surveillance, as described in chapter three, involved individualized attention, where a small number of law enforcement or intelligence personnel would compile a dossier by identifying a target, tracking, and recording their movements and activities. The targets of older forms of surveillance were often chosen because of their group membership, such as civil rights activists targeted for their race and political activism. This form of surveillance required the identification of the target before the surveillance could take place [13].

In contrast, data-based surveillance involves the emergence of the target from the data. Massive amounts of information are collected on a wide variety of individuals and groups, and then the data is mined and analyzed to identify possible targets for more thorough scrutiny. This form of surveillance does not require the identification of the target before the data collection, instead, the target is identified after the data collection [13]. .

The new surveillance is characterized by the use of high-tech tools and data analysis to sift through vast amounts of data, often without the need for individualized attention. This shift from old to new surveillance reflects the increasing reliance on data-driven approaches to surveillance and the potential for widespread indiscriminate monitoring of individuals and groups.

Public interest lawyer, homeless advocate, and emeritus professor of law at UCLA, Gary Blasi, expresses skepticism about the idea that homelessness is solely a system engineering problem, arguing that homelessness is not a problem that can be solved through data and information alone [13].

Criticism

edit

Automating Inequality: How High-Tech Tools Profile, Police, and Punish The Poor for the most part, received positive reviews from academia, media institutions and authors. Astra Taylor, author of The People’s Platform described the book as “the single most important book about technology you will read" while Dorothy Roberts, called it “a must -read” [15].

One major criticism of the book is that Eubanks did not fully capture the role race plays in poverty and discrimination by not highlighting the systemic racism that has contributed deeply to the unfair treatment and stigmatization of the poor [16].

Awards

edit

Winner: The 2019 Lillian Smith Book Award, 2018 Mc Gannon Center Book Prize and shortlisted for the Goddard Riverside Stephan Russo Book Prize for Social Justice [15].

References

edit
  1. ^ Featherstone, Liza. "How Big Data Is 'Automating Inequality'". NYTimes.com. The New York Times. Retrieved April 9, 2024.
  2. ^ Law, Tuulia (October 12, 2018). "Review of Eubanks' Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor". Surveillance & Society. 16 (3): 384–385. doi:10.24908/ss.v16i3.12612. ISSN 1477-7487.
  3. ^ Bevan, Jillian (March 29, 2020). "Eubanks, Virginia, Automating Inequality". Canadian Journal of Sociology. 45 (1): 91–94. doi:10.29173/cjs29658.
  4. ^ a b c d e f g h i j k l m n o p q Eubanks, Virginia (2019). Automating inequality: how high-tech tools profile, police, and punish the poor (First Picador ed.). New York: Picador St. Martin's Press. ISBN 978-1-250-21578-9.
  5. ^ Breidbach, Christoph F. (January 2024). "Responsible algorithmic decision-making". Organizational Dynamics: 101031. doi:10.1016/j.orgdyn.2024.101031.
  6. ^ Intelliverse.ai. "How can algorithms aid in human decision-making processes?". www.linkedin.com. Retrieved April 8, 2024.
  7. ^ a b Hyman, Mikell (December 2019). "Computers Can't Override America's Antipathy Towards the Poor - Virginia Eubanks, Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor (New York, St. Martin's Press, 2018)". European Journal of Sociology. 60 (3): 406–411. doi:10.1017/S0003975619000225.
  8. ^ a b Mañero, Julia (April 2020). "Review of Virginia Eubanks (2018). Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor.: New York: St. Martin's Press. 272 pp. ISBN 9781250074317 (Hardcover)". Postdigital Science and Education. 2 (2): 489–493. doi:10.1007/s42438-019-00077-4.
  9. ^ Gordon, Faith (November 25, 2019). "Virginia Eubanks (2018) Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor. New York: Picador, St Martin's Press". Law, Technology and Humans: 162–164. doi:10.5204/lthj.v1i0.1386.
  10. ^ Scannell, Joshua. "Broken Windows, Broken Code". Real Life. Retrieved April 7, 2024.
  11. ^ Culhane, Dennis P. (November 2010). "Tackling homelessness in Los Angeles' Skid Row: The role of policing strategies and the spatial deconcentration of homelessness". Criminology & Public Policy. 9 (4): 851–857. doi:10.1111/j.1745-9133.2010.00675.x.
  12. ^ Cornet, Maxime (December 1, 2022). "Virginia Eubanks, Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor: St. Martin's Press, New-York, 2018, 272 p.". Sociologie du travail. 64 (4). doi:10.4000/sdt.42117.
  13. ^ a b c d e f g h Eubanks, Virginia (2019). Automating inequality: how high-tech tools profile, police, and punish the poor (First Picador ed.). New York: Picador St. Martin's Press. ISBN 978-1-250-21578-9.
  14. ^ Breidbach, Christoph F. (January 2024). "Responsible algorithmic decision-making". Organizational Dynamics: 101031. doi:10.1016/j.orgdyn.2024.101031.
  15. ^ a b Macmillan. "Automating Inequality How High-Tech Tools Profile, Police, and Punish the Poor". Macmillan. Macmillan. Retrieved April 7, 2024.
  16. ^ Bevan, Jillian (March 29, 2020). "Eubanks, Virginia, Automating Inequality". Canadian Journal of Sociology. 45 (1): 91–94. doi:10.29173/cjs29658.