At this extraordinary moment in U.S. history, the evils of racism are on full display. It’s no secret that technology has played a role in enabling racism to foment and spread. This is an ideal time to read, listen, and learn. Below are many resources — research, articles, and books — that speak to the intersection of race and bias in technology, particularly in the field of AI. These are a starting point for the education that all responsible citizens should acquire.
Gender Shades – Landmark work from Joy Buolamwini, Dr. Timnit Gebru, Dr. Helen Raynham, and Deborah Raji that examines how facial recognition systems perform on different genders and races.
Voicing Erasure – A spoken word piece that was inspired by research, led by Allison Koenecke, that demonstrates how five popular speech-recognition systems perform worst on African-American Vernacular English speakers.
AI Now’s Algorithmic Accountability Policy Toolkit – A resource from the AI Now Institute “geared toward advocates interested in understanding government use of algorithmic systems,” per the organization’s website.
NIST study evaluates effects of race, age, sex on face recognition software – A report from the National Institute of Standards and Technology (NIST), part of the U.S. Chamber of Commerce.
StereoSet: A measure of bias in language models – Work from MIT that “measures racism, sexism, and otherwise discriminatory behavior in a model, while also ensuring that the underlying language model performance remains strong.”
Discriminating systems: Gender, race, and power in AI – Research from the AI Now Institute that examines the scope and scale of the diversity crisis in AI.
The future of work in black America – A report from McKinsey that looks at how automation may be widening the wealth gap between African-American families and white families in the United States.
Advancing racial literacy in tech – Work from the Data & Society project by Dr. Jessie Daniels, Mutale Nkonde, and Dr. Darakshan Mir explains why “ethics, diversity in hiring, and implicit bias training aren’t enough” to establish real racial literacy in the tech world.
Machine bias – A Pro Publica article that exposes how predictive algorithms in the criminal justice system are biased against black people.
Technological elites, the meritocracy, and postracial myths in Silicon Valley – A book chapter in which Drs. Safiya Noble and Sarah Roberts explores “some of the ways in which discourses of Silicon Valley technocratic elites bolster investments in post-racialism as a pretext for re-consolidations of capital, in opposition to public policy commitments to end discriminatory labor practices,” per the abstract.
Some key books to read on the subject of race and technology include Algorithms of Oppression by Dr. Safiya Noble, Race After Technology by Ruha Benjamin, Technicolor: Race, Technology, and Everyday Life by Alondra Nelson, Race, Rhetoric, and Technology by Dr. Adam J. Banks, and Artificial Unintelligence: How Computers Misunderstand the World by Meredith Broussard.