Identifying Police Officers at Risk of Adverse Events

Carton, Samuel; Helsby, Jennifer; Joseph, Kenneth; Mahmud, Ayesha S.; Park, Young-Soo; Walsh, Joe; Cody, Crystal; Patterson, Cpt Estella; Haynes, Lauren; Ghani, Rayid

doi:10.1145/2939672.2939698

Cited by 55 publications

(25 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of AI to predict future trends or patterns is very popular in AI4SG contexts, from applying automated prediction to redress academic failure (Lakkaraju et al 2015), to preventing illegal policing (Carton et al 2016), and detecting corporate fraud (Zhou and Kapoor 2011). The predictive power of AI4SG faces two risks: the manipulation of input data, and excessive reliance on non-causal indicators.…”

Section: Safeguards Against the Manipulation Of Predictorsmentioning

confidence: 99%

“…Projects seeking to use AI for social good vary significantly. They range from models to predict septic shock (Henry et al 2015) to game-theoretic models to prevent poaching (Fang et al 2016); from online reinforcement learning to target HIV-education at homeless youths (Yadav et al 2016a, b) to probabilistic models to prevent harmful policing (Carton et al 2016) and support student retention (Lakkaraju et al 2015). Indeed, new applications of AI4SG appear almost daily, making possible socially good outcomes that were once less easily achievable, unfeasible, or unaffordable.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

How to Design AI for Social Good: Seven Essential Factors

et al. 2020

View full text Add to dashboard Cite

The idea of artificial intelligence for social good (henceforth AI4SG) is gaining traction within information societies in general and the AI community in particular. It has the potential to tackle social problems through the development of AI-based solutions. Yet, to date, there is only limited understanding of what makes AI socially good in theory, what counts as AI4SG in practice, and how to reproduce its initial successes in terms of policies. This article addresses this gap by identifying seven ethical factors that are essential for future AI4SG initiatives. The analysis is supported by 27 case examples of AI4SG projects. Some of these factors are almost entirely novel to AI, while the significance of other factors is heightened by the use of AI. From each of these factors, corresponding best practices are formulated which, subject to context and balance, may serve as preliminary guidelines to ensure that well-designed AI is more likely to serve the social good.

show abstract

Section: Safeguards Against the Manipulation Of Predictorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

How to Design AI for Social Good: Seven Essential Factors

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Many notable cases of algorithmic biases involve the algorithm mirroring structural inequities and biases that are unfortunately widespread in society. In fact, there are even recent efforts to use the presence of algorithmic bias as a ‘signal’ of previously unnoticed real‐world biases that should potentially be ameliorated (Carton et al., 2016; Pierson et al., 2020).…”

Section: Sources Of Biasmentioning

confidence: 99%

Algorithmic bias: Senses, sources, solutions

Fazelpour

Danks

2021

Philosophy Compass

View full text Add to dashboard Cite

Data‐driven algorithms are widely used to make or assist decisions in sensitive domains, including healthcare, social services, education, hiring, and criminal justice. In various cases, such algorithms have preserved or even exacerbated biases against vulnerable communities, sparking a vibrant field of research focused on so‐called algorithmic biases. This research includes work on identification, diagnosis, and response to biases in algorithm‐based decision‐making. This paper aims to facilitate the application of philosophical analysis to these contested issues by providing an overview of three key topics: What is algorithmic bias? Why and how can it occur? What can and should be done about it? Throughout, we highlight connections—both actual and potential—with philosophical ideas and concerns.

show abstract

“…In settings in which there are rich cross-sectional data-for example, detailed demographic data or pre-employment information-predictions about police officer risk are typically made using sophisticated machine learning-based algorithms (Carton et al, 2016;Chalfin et al, 2016;Helsby et al, 2018;Ridgeway & MacDonald, 2009) or, at a minimum, logistic regression (Leinfelt, 2005;White, 2008). 13 The advantage of machine learning methods in such a context is that the approach allows researchers to automate the detection of signal in the data, a task which is complicated considerably when the number of predictors is large and the relationships between variables are nonlinear and conditional (Hastie et al, 2009).…”

Section: Persistence In Complaintsmentioning

confidence: 99%

How many complaints against police officers can be abated by incapacitating a few “bad apples?”

Chalfin

Kaplan

2021

Criminology & Public Policy

View full text Add to dashboard Cite

The notion that the unjustified use of force by police officers is concentrated among a few "bad apples" is a popular descriptor that has gained traction in scholarly research and achieved considerable influence among policy makers. But is removing the bad apples likely to have an appreciable effect on police misconduct? Leveraging a simple policy simulation and data from the Chicago Police Department, we estimate that removing the top 10% of officers identified based on ex ante risk and replacing them with officers drawn from the middle of the risk distribution would have led to only a 4-6% reduction in the use of force incidents in Chicago over a 10-year period. Policy Implications: Our analysis suggests that surgically removing predictably problematic police officers is unlikely to have a large impact on citizen complaints.By assembling some of the first empirical evidence on the likely magnitude of incapacitation effects, we provide critical support for the idea that early warning systems must be designed, above all, to deter problematic behavior and promote accountability.

show abstract

Identifying Police Officers at Risk of Adverse Events

Cited by 55 publications

References 9 publications

How to Design AI for Social Good: Seven Essential Factors

How to Design AI for Social Good: Seven Essential Factors

Algorithmic bias: Senses, sources, solutions

How many complaints against police officers can be abated by incapacitating a few “bad apples?”

Contact Info

Product

Resources

About