In the context of regression, we consider the fundamental question of making an estimator fair while preserving its prediction accuracy as much as possible. To that end, we define its projection to fairness as its closest fair estimator in a sense that reflects prediction accuracy. Our methodology leverages tools from optimal transport to construct efficiently the projection to fairness of any given estimator as a simple post-processing step. Moreover, our approach precisely quantifies the cost of fairness, measured in terms of prediction accuracy.

Many critics of police facial recognition use warn of the potential for racial bias that leads to false arrests, and unfortunately that appears to have happened. The ACLU has filed a complaint against Detroit police for the wrongful arrest of Robert Williams when a DataWorks Plus facial recognition system incorrectly matched security footage against Williams' driver's license, marking him as a suspect. Officers showed the match to an offsite security consultant who identified Williams as the culprit, but this person never saw the perpetrator first-hand. The ACLU argued that the DataWorks system "can't tell Black people apart" and that the whole system was "tainted" by officers' assumptions that the facial recognition system produced the right suspect. In a Washington Post opinion piece, Williams added that he was concerned about the tech even if it was completely accurate -- he didn't want his daughters' faces to go into a database and prompt future police questioning when they're spotted at a "protest the government didn't like."

Over 1500 researchers across multiple fields have banded together to openly reject the use of technology to predict crime, arguing it would reproduce injustices and cause real harm. The Coaltition for Critical Technology wrote an open letter to Springer Verlag in Germany to express their grave concerns about a newly developed automated facial recognition software that a group of scientistts from Harrisburg Univeristy, Pennsylvania have developed. Springer's Nature Research Book Series intends to publish an article by the Harrisburg scientists named A Deep Neural Network Model to Predict Criminality Using Image Processing. The coalition wants the publication of the study - and others in similar vein - to be rescinded, arguing the paper makes claims that are based unsound scientific premises, research and methods. Developed by a New York Police Department veteran and PhD student Jonathan Korn along with professors Nathaniel Ashby and Roozbeh Sadeghian, the Harrisburg University researchers' software claims 80 per cent accuracy and no racial bias.

A photo of the alleged suspect in a theft case in Detroit, left, next to the driver's license photo of Robert Williams. An algorithm said Williams was the suspect, but he and his lawyers say the tool produced a false hit. A photo of the alleged suspect in a theft case in Detroit, left, next to the driver's license photo of Robert Williams. An algorithm said Williams was the suspect, but he and his lawyers say the tool produced a false hit. Police in Detroit were trying to figure out who stole five watches from a Shinola retail store.

Robert Williams spent over a day in custody in January after face recognition software matched his driver's license photo to surveillance video of someone shoplifting, the American Civil Liberties Union of Michigan (ACLU) said in the complaint. In a video shared by ACLU, Williams says officers released him after acknowledging "the computer" must have been wrong.

The ACLU complaint said Detroit police "unthinkingly relied on flawed and racist facial recognition technology without taking reasonable measures to verify the information being provided." It called the resulting investigation "shoddy and incomplete," the officers involved "rude and threatening," and said the department has dragged its feet responding to public-information requests for relevant records.

Quota-based fairness mechanisms like the so-called Rooney rule or four-fifths rule are used in selection problems such as hiring or college admission to reduce inequalities based on sensitive demographic attributes. These mechanisms are often viewed as introducing a trade-off between selection fairness and utility. In recent work, however, Kleinberg and Raghavan showed that, in the presence of implicit bias in estimating candidates' quality, the Rooney rule can increase the utility of the selection process. We argue that even in the absence of implicit bias, the estimates of candidates' quality from different groups may differ in another fundamental way, namely, in their variance. We term this phenomenon implicit variance and we ask: can fairness mechanisms be beneficial to the utility of a selection process in the presence of implicit variance (even in the absence of implicit bias)? To answer this question, we propose a simple model in which candidates have a true latent quality that is drawn from a group-independent normal distribution. To make the selection, a decision maker receives an unbiased estimate of the quality of each candidate, with normal noise, but whose variance depends on the candidate's group. We then compare the utility obtained by imposing a fairness mechanism that we term $\gamma$-rule (it includes demographic parity and the four-fifths rule as special cases), to that of a group-oblivious selection algorithm that picks the candidates with the highest estimated quality independently of their group. Our main result shows that the demographic parity mechanism always increases the selection utility, while any $\gamma$-rule weakly increases it. We extend our model to a two-stage selection process where the true quality is observed at the second stage. We discuss multiple extensions of our results, in particular to different distributions of the true latent quality.

Last week, the Supreme Court issued a surprising 6–3 decision barring hiring discrimination against LGBTQ people under Title VII of the Civil Rights Act, with conservative Justice Neil Gorsuch making the textualist case for this landmark protection. The unexpected outcome in Bostock v. Clayton County should provoke introspection among progressives in the legal community who have long been skeptical of textualism, offering a chance for them to fix chronic blind spots and strategic gaffes that have damaged the progressive judicial project. While it's clear that this ruling was a major victory for progressives, less apparent is how, going forward, progressive advocates, judges, and politicians should think and talk about statutory interpretation. Although brow-furrowing, that question is hugely important. As the late high priest of conservative textualism, Justice Antonin Scalia, pointed out: "By far the greatest part of what I and all federal judges do is interpret the meaning of federal statutes."

In a new letter, more than 1,600 Google workers are demanding that the company end its work with police departments across the country. "The past weeks have shown us that addressing racism is not merely an issue of words, but of actions taken to dismantle the actual structures that perpetuate it," the workers wrote in a draft of the to-be-released letter addressed to Google CEO Sundar Pichai and obtained by Mother Jones. The letter was signed by 1,670 employees, according to a screenshot that was shared with Mother Jones by someone with access to the signature list, and was organized by Googlers Against Racism, an advocacy group within the company. "While we as individuals hold difficult but necessary conversations with our family, friends and peers, we are also incredibly disappointed by our company's response," the letter continued, referencing Google's lip service to the Black Lives Matter movement. The letter also demands that the company "stop making our technology available to police forces."

The Coalition for Critical Technology (CCT) penned a letter opposing the publication of research called "A Deep Neural Network Model to Predict Criminality Using Image Processing." At the time of publication the letter has more than 1,000 signatures from researchers, practitioners, academics, and others. According to a press release from Harrisburg University, the paper is slated for publication in a book series from Springer Publishing, and the letter urges readers to demand that Springer pull the paper and condemn the use of criminal justice statistics to predict criminality. The use of algorithms in predictive policing is a fraught subject. As the CCT letter elaborates, criminal justice data is notoriously flawed.