Algorithms: researchers accuse technology of intensifying structural racism

RIO DE JANEIRO, BRAZIL – A black person who is automatically recognized as a gorilla on a digital photo platform. On one social media, the automatic cropping of an off-viewing photo privileges white people’s faces. On another network, a black woman has her post reach increased by 6,000% by posting white women.

These examples are not isolated and have been the target of criticism and reflection by internet users and researchers. How could mathematical models, the so-called algorithms, be racist?

Researcher Tarcizio Silva, PhD student in Human and Social Sciences at the Federal University of ABC (UFABC), explains that it is necessary to ask how these systems are used in a way that allows “the maintenance, intensification, and concealment of structural racism”. Silva has developed a timeline that demonstrates cases, data, and reactions.

“The solution lies not only in code transparency but in the appropriation and social critique of technology,” he says. How the systems are fed, what data is accepted, who creates the technologies, and who is included or excluded in the multiplication of automated devices are some of Silva’s questions. “Algorithmic racism is a technologization and automation of structural racism,” he evaluates.

Idealizers of the Tecnocríticas blog, Renata Gusmão, Gabriela Guerra, and Felipe Martins, work in information technology (IT) and use the Internet to discuss other issues involving the lack of neutrality of technology.

“The people who think up these algorithms are people within a sexist, racist, and unequal society. Soon the logic behind a solution carries these same values. They are not considering the diversity of the final users and end up reinforcing inequalities and discriminations of the ‘real’ world,” they point out in an e-mail interview with Agência Brasil.

#RacistAlgorithm

One of the cases of greatest repercussion recently occurred on Twitter, with the automatic cropping of photos that favored white faces. Thousands of users used the hashtag #AlgoritmoRacista, on the network itself, to question the automation that exposed racism. Silva explains that this discovery showed algorithms based on neural networks, whose technique finds regions of interest on the image from data gathered by gaze tracking.

“An accumulation of data and biased research that privileged white aesthetics resulted in the system that Twitter used and failed to even properly explain where the origin of the issue was,” the researcher said. At the time, the platform pledged to revise the mechanism.

“We should have done a better job of foreseeing this possibility when we were designing and building this product.”

“This is how algorithmic racism works, through the accumulation of using poorly explained and poorly tuned technologies that at first optimize some technical aspect, but actually mutilate the user experience,” the researcher adds.

Facial Recognition

Outside of social networks, the damage of algorithmic racism can be even greater. Data surveyed by the Network of Security Observatories show that from March to October 2019, 151 people were arrested from technology and facial recognition in four states (Bahia, Rio de Janeiro, Santa Catarina, and Paraíba).

Records that had information about race and color, or when there were images of the people approached (42 cases), observed that 90.5% were black. “The main motivations for the approaches and arrests were drug trafficking and robbery,” points out the report.

Silva recalls that this technology has been the target of questioning or banning in countries in Europe and regions of the United States. “The reasons are various, from inaccuracy to low cost-benefit or the promotion of vigilantism and state violence,” he explains.

He points out that the system is inaccurate in identifying minority faces. “But never mind a future where facial recognition is more accurate: it is a necessarily racist technology in countries where penal selectivity and mass incarceration are the modus operandi of the state.”

Exits

For the members of Tecnocríticas, fighting this discriminatory expression of algorithms involves ensuring more diversity in the IT area.

“Whether by guaranteeing that the teams responsible for thinking up these solutions have racial and gender diversity, for example or by training robots with diverse data. Another issue, also critical, is that the technology industry responds to economic dynamics, so these solutions are not always the ones that really solve people’s problems. Still, the ones that generate profit,” they evaluate.

Silva believes that the first step towards protection is to “overcome any presumption of neutrality of technologies. He points out that emerging digital technologies, such as facial recognition for police purposes or risk score creation for private health plans, are already “born as humanitarian defeats.”

“If we effectively commit to principles of the value of human life, we will conclude that some algorithmic technologies should not even exist.”

Source: Agência Brasil