Kristian Lum’s work on the HRDAG Policing Project is referred to here: “In fact, Lum argues, it’s not clear how well this model worked at depicting the situation in Oakland. Those data on drug crimes were biased, she now reports. The problem was not deliberate, she says. Rather, data collectors just missed some criminals and crime sites. So data on them never made it into her model.”

Kristian Lum discusses the challenges of getting accurate data from conflict zones, as well as her concerns about predictive policing if law enforcement gets it wrong.

In her work on statistical issues in criminal justice, Lum has studied uses of predictive policing—machine learning models to predict who will commit future crime or where it will occur. In her work, she has demonstrated that if the training data encodes historical patterns of racially disparate enforcement, predictions from software trained with this data will reinforce and—in some cases—amplify this bias. She also currently works on statistical issues related to criminal “risk assessment” models used to inform judicial decision-making. As part of this thread, she has developed statistical methods for removing sensitive information from training data, guaranteeing “fair” predictions with respect to sensitive variables such as race and gender. Lum is active in the fairness, accountability, and transparency (FAT) community and serves on the steering committee of FAT, a conference that brings together researchers and practitioners interested in fairness, accountability, and transparency in socio-technical systems.

“Every kind of classification system – human or machine – has several kinds of errors it might make,” [Patrick Ball] says. “To frame that in a machine learning context, what kind of error do we want the machine to make?” HRDAG’s work on predictive policing shows that “predictive policing” finds patterns in police records, not patterns in occurrence of crime.

Patrick Ball, estadístico norteamericano, colabora con el Programa de Derechos Humanos de la Universidad Iberoamericana en una investigación sobre fosas clandestinas.

El modelo “puede distinguir entre los municipios en que vamos a encontrar fosas clandestinas, y en los que es improbable que vayamos a encontrar estas fosas”, explicó Patrick Ball, estadístico estadounidense que colabora con el Programa de Derechos Humanos de la Universidad Iberoamericana de la Ciudad de México.

When Patrick Ball was introduced to Ibero’s database, the director of research at the Human Rights Data Analysis Group in San Francisco, California, saw an opportunity to turn the data into a predictive model. Ball, who has used similar models to document human rights violations from Syria to Guatemala, soon invited Data Cívica, a Mexico City–based nonprofit that creates tools for analyzing data, to join the project.

Por ello, Human Rights Data Analysis Group (HRDAG), el Programa de Derechos Humanos de la Universidad Iberoamericana (UIA) y Data Cívica, realizan un análisis estadístico construido a partir de una variable en la que se identifican fosas clandestinas a partir de búsquedas automatizadas en medios locales y nacionales, y usando datos geográficos y sociodemográficos.

“If you’re not careful, you risk automating the exact same biases these programs are supposed to eliminate,” says Kristian Lum, the lead statistician at the San Francisco-based, non-profit Human Rights Data Analysis Group (HRDAG). Last year, Lum and a co-author showed that PredPol, a program for police departments that predicts hotspots where future crime might occur, could potentially get stuck in a feedback loop of over-policing majority black and brown neighbourhoods. The program was “learning” from previous crime reports. For Samuel Sinyangwe, a justice activist and policy researcher, this kind of approach is “especially nefarious” because police can say: “We’re not being biased, we’re just doing what the math tells us.” And the public perception might be that the algorithms are impartial.

Kristian Lum, lead statistician at the Human Rights Data Analysis Group (and letter signatory), fears that “in order to flag even a small proportion of future terrorists, this tool will likely flag a huge number of people who would never go on to be terrorists,” and that “these ‘false positives’ will be real people who would never have gone on to commit criminal acts but will suffer the consequences of being flagged just the same.”

HRDAG analyst William Isaac is quoted in this article about so-called crime prediction. “They’re not predicting the future. What they’re actually predicting is where the next recorded police observations are going to occur.”