Firstly, let me acknowledge that I am no expert in the megatrend of machine learning and therefore writing about an atypical organization might be a bit of a bold move. I shall express my thoughts nonetheless \u2013 because of how much this topic area has intrigued me over the past few days.<\/p>\n
My choice of organization for this analysis is the US Justice Department. Machine Learning as a megatrend has been extremely critical to the \u201cprocess improvement\u201d for global justice systems at large, and in this case the extent of adoption and its implications can be studied by diving deeper into the evolution of trials in some US courts. What makes machine-learning adoption (or the lack of it) as critical in this context is the gravity of its implications \u2013 we\u2019re juggling between inherent human biases and potentially faulty algorithms to decide on the future of defendants in the court of law.<\/p>\n
Several studies in the past, including Austin & Williams\u2019 1977 note on sentencing disparity<\/em> [1], have highlighted the consistency problem with the judicial system \u2013 bringing to light the fact that court judges even in the same state end up with very disparate judgements for the same cases. In more than a few instances, it has also been highlighted that the same judges arrive at different judgements when presented with the exact same case, just at a different point in time [2]. \u00a0This makes it apparent that there are several biases at play when forming judgement, as with every other scenario that requires human interpretation.<\/p>\n With the courtroom stakes as high as they are, it would make sense that the US Justice Department would seek to employ methods that could help eliminate some of these human biases. Therefore, to no surprise, the Justice Department\u2019s National Institute of Corrections encourages the use of \u201c(algorithm-based) assessments at every stage of the criminal justice process\u201d [3]. After all, using algorithms in the process allows for greater consistency \u2013 with the expectation that the outcome will be more standardized and therefore equitable for every case where the parameters match. However, my research in this area has highlighted the shortcomings of this very claim \u2013 because it entirely depends on comparing cases on a pre-defined set of parameters, ignoring any other context the case may provide.<\/p>\n While there are several dimensions to this challenge, we shall explore the risks of overdependence on algorithms using one simple real-world example. Privately developed tools such as COMPAS<\/em> [4] \u2013 which have been designed to assess risk and predict re-offense rates of defendants \u2013 have been commonly used across several state courts to make decisions on plea bargains. The deployment of this tool in Wisconsin\u2019s court was subject to negative publicity when in one case (Paul Zilly \u2013 convicted of stealing a lawnmower), the judges changed their decision from agreeing to a plea deal to instead doubling Zilly\u2019s sentence to two years in state prison, based on the risk assessment presented by the algorithm [3]. While the algorithm is \u201cproprietary\u201d and there is no way of knowing the true accuracy of the prediction, what is apparent is that there is a bias towards believing that the \u201cmachine\u201d is right \u2013 therefore putting additional onus on the computer program being fair and holistic.<\/p>\n Are the systems that are being used fair and holistic however? Are our machine-learning capabilities sophisticated enough that we value system-generated trial judgements over the \u201csubjective\u201d conclusions of judges who come with decades of experience and the ability to empathize with individual contexts? While there is no single answer to these questions, my analysis leads me to believe that there are sufficient grounds to be skeptical.<\/p>\n There is a school of thought with the belief that the increased dependence on algorithms will allow us to eliminate human bias during trials, which as I understand is not necessarily true. At the very core of this challenge is that even in very sophisticated programs, the \u2018machine\u2019 is set up to \u2018learn\u2019, at least partially, from historic incidences and behavior of crime. One of the several implications of this phenomena is that the algorithms will continue to make race and gender biased predictions, even when race and gender might not be coded as parameters to match against, leading to increased false positives <\/em>[5].<\/p>\n It is important therefore, especially in a sensitive and high-stake environment such as the judicial system, to \u201cget it right\u201d. What should be the stance of the US Justice Department in this transitionary phase of machine learning sophistication? How do we strike a balance between taking advantage of the technology currently available and the traditional procedure for trials? Should all computer programs in this field go through a centralized audit and review process before being allowed to go-to-market? Looking forward to your comments!<\/p>\n (Word Count: 783)<\/p>\n <\/p>\n Endnotes:<\/strong><\/p>\n Overall concept credit: [1] William Austin and Thomas A. III Williams, “A Survey of Judges’ Responses to Simulated Legal Cases: Research Note on Sentencing Disparity”, The Journal of Criminal Law and Criminology, Vol. 68 Issue 2,\u00a0https:\/\/scholarlycommons.law.northwestern.edu\/jclc\/vol68\/iss2\/12\/<\/a>\u00a0accessed November 2018<\/p>\n [2] Ben Bryant. “Judges are more lenient after taking a break, study finds”, The Guardian, April 11, 2011.\u00a0https:\/\/www.theguardian.com\/law\/2011\/apr\/11\/judges-lenient-break<\/a>\u00a0accessed November 2018<\/p>\n [3] Julia Angwin, Jeff Larson, Surya Mattu and Lauren Kirchner, “Machine Bias”, Pro Publica, May 23, 2016.\u00a0\u00a0https:\/\/www.propublica.org\/article\/machine-bias-risk-assessments-in-criminal-sentencing<\/a>\u00a0accessed November 2018<\/p>\n
\nHannah Fry. Hello World: Being Human in the Age of Algorithms.<\/em> W. W. Norton & Company Inc. (2018)
\n<\/em>(where the author provides a narrative on the evolution, current applications and potential future of algorithms across seven industries)<\/p>\n