Queen Carrasco 

10 May 2023

Writing For Engineering

The New Jim Code: Understanding Bias In Tech

 

Introduction:

 My name is Queen Carrasco, and I am a first-year student in the CUNYBA Program, studying Social Data Analytics. I started my time here majoring in Computer Science and minoring in Sociology. However, I realized I wanted to combine the two due to my interests relating to activism, and the impact of data on the world – specifically minority communities. How can data shape us? How can data tell a story? How can data influence us? One such way data can influence us is through the New Jim Code. 

 

Relevant Terms:

 

Deep learning is the subfield of machine learning in which ‘depth’ refers to the layers of abstraction that a computer program makes, learning more complicated concepts by building them out of simpler ones. The New Jim Code is thought of as the employment of new technologies that reflect and reproduce existing inequities but that are promoted and perceived as more objective or progressive than the discriminatory systems of a previous era. It was coined by Ruha Benjamin, sociologist and professor of African American studies at Princeton University. Author of three books, including Race After Technology (2019), she writes, teaches, and speaks about the relationship between innovation and inequity, knowledge and power, race and citizenship, and health and race.

 

What are the origins of “the New Jim Code”?

 

The term Jim Crow originated in 1832, and was the name of the title character of a minstrel show. Eventually, it evolved into a racial epithet and term used for legalized segregated spaces. The New Jim Crow is a book written by Michelle Alexander in 2012; Alexander’s book made the case that the expansion of the prison industrial complex was creating a new racial caste system via the use of “colorblind” ideology within the criminal justice system In 2019, Benjamin continued with Alexander’s idea, with the New Jim Code. Benjamin emphasized that “colorblind” technology is really organizing and facilitating social control via discriminatory practices.

 

Where can “the New Jim Code” be seen?

Embedded in artificial intelligence, search algorithms, facial recognition software, and even motion detection, Jim Code can be seen. Joy Boulamwini, a computer scientist and digital activist at MIT Media Lab, conducted a study in 2018. The study compared the facial recognition software from IBM, Microsoft, and Face++ to see which software correctly identified the faces presented to it. The study encompassed both race and gender of the subjects. All classifiers performed better on male faces than female faces (8.1% – 20.6% difference in error rate); all classifiers performed better on lighter faces than darker faces (11.8% – 19.2% difference in error rate); all classifiers performed worst on darker female faces (20.8% – 34.7% error rate). Overall, Microsoft performed the best with an error rate of 0.7% for lighter skin faces and 12.9% for darker skin faces. IBM performed the worst with an error rate of 22.4% for darker skin faces.

In 2016, a study of recidivism scores assigned to thousands arrested in Broward County found that the risk assessment software used by the police was assigning a higher risk assessment for black prisoners compared to white prisoners. One such case was with Dylan Fugett, a white prisoner, and Bernard Parker, a black prisoner. At the time, Fugett was charged with a count of attempted burglary, while Parker was charged with nonviolently resisting arrest. Fugett was rated a 3, while Parker was rated a 10. However, Fugett was the one to be arrested once more for drug possession. Parker was never arrested again. 

 

How can we combat “the New Jim Code”?

We can combat the New Jim Code in two ways:

1. Rethink the ways in which we design our tools – By modifying the ways in which we create our algorithms or develop our software, we can highlight the human being under the microscope. However, this step takes time and money. Additionally, racism cannot be dismantled in one day. It is one step at a time.
2. Create effective policy solutions – Policies create a pathway for safe and more equitable communities. Additionally, the actions are more immediate – as well as have a long-lasting effect.

 

References: 

 

1. Benjamin, R. (2019). Race after technology: Abolitionist tools for the new Jim code. Polity.
2. Brownlee, J. (2019, August 16). What is Deep Learning? – MachineLearningMastery.com. Machine Learning Mastery. Retrieved May, 7 2023 from https://machinelearningmastery.com/what-is-deep-learning
3. Buolamwini, J. &; Gebru, T.. (2018). Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification. Proceedings of the 1st Conference on Fairness, Accountability and Transparency, in Proceedings of Machine Learning Research 81:77-91 Available from https://proceedings.mlr.press/v81/buolamwini18a.html.
4. Ferguson, A. G. (2015). BIG DATA AND PREDICTIVE REASONABLE SUSPICION. University of Pennsylvania Law Review, 163(2), 327–410. http://www.jstor.org/stable/24247848    
5. Mathews, A. S. (2022, February 14). Scholar discusses solutions to New Jim Code. Technique. Retrieved May 5, 2023, from https://nique.net/life/2022/02/14/scholar-discusses-solutions-to-new-jim-code/
6. Smith, B. (2020, March 31). Finally, progress on regulating facial recognition – Microsoft On the Issues. The Official Microsoft Blog. Retrieved May 8, 2023, from https://blogs.microsoft.com/on-the-issues/2020/03/31/washington-facial-recognition-legislation/
7. Varghese, S. (2019, June 29). Ruha Benjamin: ‘We definitely can’t wait for Silicon Valley to become more diverse’. The Guardian. Retrieved May 8, 2023, from https://www.theguardian.com/technology/2019/jun/29/ruha-benjamin-we-cant-wait-silicon-valley-become-more-diverse-prejudice-algorithms-data-new-jim-code