The Odds of Justice: Code of Silence: <i>How private companies hide flaws in the software that governments use to decide who goes to prison and who gets out</i>

Wexler, Rebecca

doi:10.1080/09332480.2018.1522217

Cited by 13 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since we do not know COMPAS, we cannot check this possibility. In the past, there have been documented cases where individuals have received incorrect COMPAS scores based on incorrect criminal history data (16,17), and no mechanism to correct it after a decision was made based on that incorrect score. We do not know whether this happens often enough to influence the scatter plot in a visible way.…”

Section: Caveatsmentioning

confidence: 99%

“…In cases like that of prisoner Glenn Rodrïguez whose parole was denied because of a miscalculated COMPAS score (16,17), he did not notice the error on his COMPAS form until after his…”

Section: Compas Sometimes Labels Individuals With Long or Serious Crimentioning

confidence: 99%

“…In cases like that of prisoner Glenn Rodrïguez whose parole was denied because of a miscalculated COMPAS score (16,17), he did not notice the error on his COMPAS form until after his parole was denied. Complicated forms, even if one is asked to check them over, lead to human error.…”

Section: Compas Sometimes Labels Individuals With Long or Serious Cri...mentioning

confidence: 99%

“…Individuals with long criminal histories are sometimes given low COMPAS scores (which labels them as low risk), and vice versa. In the past, there have been documented cases where individuals have received incorrect COMPAS scores based on incorrect criminal history data (16,17) and have had no mechanism to correct it after a decision was made based on that incorrect score. This problem has inadvertently occurred with other (even transparent) scoring systems, at least in one case leading to the release of a dangerous individual who committed a murder while on bail (18,19).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The Age of Secrecy and Unfairness in Recidivism Prediction

Rudin¹,

Wang²,

Coker³

2020

Harvard Data Science Review

View full text Add to dashboard Cite

and ultimately stagnating because there is no clear definition of fairness and competing definitions are largely incompatible. We argue that the focus on the question of fairness is misplaced, as these algorithms fail to meet a more important and yet readily obtainable goal: transparency. As a result, creators of secret algorithms can provide incomplete or misleading descriptions about how their models work, and various other kinds of errors can easily go unnoticed.By trying to partially reconstruct the COMPAS model -a recidivism-risk scoring model used throughout the criminal justice system -we show that it does not seem to depend linearly on the defendant's age, despite statements to the contrary by the model's creator. This observation has not been made before despite many recently published papers on this algorithm. Furthermore, by subtracting from COMPAS its (hypothesized) nonlinear age component, we show that COMPAS does not necessarily depend on race, contradicting ProPublica's analysis, which assumed linearity in age. In other words, faulty assumptions about a proprietary algorithm lead to faulty conclusions that go unchecked. Were the algorithm transparent in the first place, this likely would

show abstract

Section: Caveatsmentioning

confidence: 99%

“…In cases like that of prisoner Glenn Rodrïguez whose parole was denied because of a miscalculated COMPAS score (16,17), he did not notice the error on his COMPAS form until after his…”

Section: Compas Sometimes Labels Individuals With Long or Serious Crimentioning

confidence: 99%

Section: Compas Sometimes Labels Individuals With Long or Serious Cri...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The Age of Secrecy and Unfairness in Recidivism Prediction

Rudin¹,

Wang²,

Coker³

2020

Harvard Data Science Review

View full text Add to dashboard Cite

show abstract

“…This eventuality urges us to extend the current debate on how AI agents contribute to shaping societal reality (Wexler, 2017), particularly when thinking about the current media landscape through selective exposure of individuals to partisan views that increase polarisation and populist discourse.…”

Section: The Mediation Problem: Direct Communication Between Stakehol...mentioning

confidence: 99%

Ethical implications of text generation in the age of artificial intelligence

Illia

Colleoni

Zyglidopoulos

2022

Business Ethics Env & Resp

View full text Add to dashboard Cite

We are at a turning point in the debate on the ethics of Artificial Intelligence (AI) because we are witnessing the rise of general-purpose AI text agents such as GPT-3 that can generate large-scale highly refined content that appears to have been written by a human. Yet, a discussion on the ethical issues related to the blurring of the roles between humans and machines in the production of content in the business arena is lacking. In this conceptual paper, drawing on agenda setting theory and stakeholder theory, we challenge the current debate on the ethics of AI and aim to stimulate studies that develop research around three new challenges of AI text agents: automated mass manipulation and disinformation (i.e., fake agenda problem), massive low-quality content production (i.e., lowest denominator problem) and the creation of a growing buffer in the communication between stakeholders (i.e., the mediation problem). | INTRODUC TI ONThe rise of AI agents for automated text generation (ATG) brings with it new ethical challenges. Such agents can generate large-scale highly refined content that 'sounds' like a real human, and their use is on the rise. This development has profound implications for society:What happens if we end up silencing the human voice? Who benefits from this technology and who loses out? And how should we regulate its use to ensure responsible deployment in public discourse? These questions are addressed through an analysis of current developments in AI agent technologies for ATG. This paper identifies three core issues: (1) there is no consensus yet as to what constitutes good or bad ATG; (2) social media platforms have not yet developed policies on ATG; and (3) existing regulation does not deal sufficiently with risks posed by ATG. It concludes that more research needs to be done into different types of text generated by AI agents before any universal rules about appropriate use can be established. In addition, it argues that social media companies need to develop policies regarding acceptable uses of such software, while governments need to review existing legislation so as to ensure proper oversight over their deployment. The first paragraph might sound like a well-crafted introduction written by academic scholars. Yet, this text was entirely generated by GPT-3 (Generative Pre-trained Transformer), the world's most advanced AI for ATG. To generate this text, we simply gave GPT-3 the following instructions: 'This is the abstract of a scientific article in the academic Journal Business Ethics, the Environment and Responsibility that discusses ethical challenges posed by the end of the era of rule-based bots and the rise of easy-to-use, thirdgeneration, general-purpose, Artificial Intelligence (AI) agents for ATG that can generate large-scale highly refined content that reads like a human. The article focuses on the risk of massive production of text by AI agents and the implications of silencing human voices. The research question that is posed is…' 1 and asked GPT-3 to produce a text 288 chara...

show abstract

Emerging Futures and Technology Ethics

Sage

Griffin

2022

Social Work Theory and Ethics

View full text Add to dashboard Cite

The Odds of Justice: Code of Silence: How private companies hide flaws in the software that governments use to decide who goes to prison and who gets out

Cited by 13 publications

References 0 publications

The Age of Secrecy and Unfairness in Recidivism Prediction

The Age of Secrecy and Unfairness in Recidivism Prediction

Ethical implications of text generation in the age of artificial intelligence

Emerging Futures and Technology Ethics

Contact Info

Product

Resources

About