Prejudice in the tech industry is spilling into technologies

3 min read

prejudice-tech-industryBy Desmond Rhodes

Technology, by nature, cannot be discriminatory. It can’t hold prejudice beliefs against anyone, and won’t treat anyone differently because of their color, race, creed, sexual orientation, socioeconomic status, etc. However, many people have claimed that some technologies are discriminatory, and it has shown to be true. Facial recognition programs and dating apps have shown to be built with algorithmic biases. However, could it be that the people making these technologies hold biased beliefs?

An important distinction needs to be made when people say that some technologies are discriminatory. The technology itself isn’t prejudiced; it’s the people programming these technologies that may hold racist and sexist preconceptions. Before this article gets started, it’s good to point out that the people who program these technologies are probably not actively racist. It would be more plausible to consider the fact that the tech industry predominantly consists of white men who gear these technologies the way they understand the world and not taking people of color into account when entering extensive data into predictive technologies such as AI, machine, and deep learning.

How can this be? Read below to find out some gender and racial disparities happening in the tech industry, and how it could explain racial and sexist connotations of some technologies.

gender-gap

The gender gap

You may be aware that there is a gender gap in the tech industry, but you may not know the magnitude of the situation. The tech industry has infamously adopted the term “tech bro culture,” equating the goings on in tech sectors to that of a frat house; women in the tech industry have endured immature boys, bullying, sexism, and even reports of sexual harassment.

Many cases of unwanted sexual advances and other sexual harassment have been swept under the rug, perpetuating the volatile culture in the tech industry. However, women continue to fight back. One of the leading female figures battling against sexual harassment in the tech industry is Ellen Pao. Business Insider reported that:

“In 2012, VC Ellen Pao famously sued Kleiner Perkins alleging sexual discrimination, not harassment. But in the trial, she alleged that one of her co-workers tried to retaliate after she ended an affair with him. She ultimately lost the case. That partner, Ajit Nazre, left the job and was accused of sexual harassment by another female VC at the firm.”

Ellen Pao is just one of the many women that have been discouraged; many others have considered leaving the tech field for a new career. As a result, the tech industry is becoming more and more male-dominated. This gender gap is only the beginning of males isolating themselves in the tech sector — that is, until more tech companies like Alibaba start advocating for women in the tech field.

Racial disparities

Being a woman in the tech industry is tough, and being a non-white woman in the tech industry is even more so. Minorities — men and women — are finding that they are running into a brick wall when trying to make a name for themselves or simply even breaking into the tech field.

City Lab sheds light on disparities in the tech industry when they state, “white women were 31 percent more likely than Hispanic men to be executives, and 88 percent and 97 percent more likely than Asian and black men respectively. Meanwhile, for minority women, the ‘race-to-gender factor’ has only worsened since 2007.” Taking into account the adversities women face when falling victim to the gender gap, City Lab shows that the racial divide is even more significant.

If tech companies are willing, even if reluctantly, to hire white women, they will be even more hesitant to hire Hispanic, Asian, and Black men. Just imagine the hardships of a woman who is one of these races; it could prove to be almost impossible to break into tech.

The gender gap, combined with racial disparities, is why white males dominate the tech industry. A white male mindset may explain why when programming predictive technologies and algorithms, minorities aren’t taken into account. This mindset is making for some embarrassing racial assumptions of particular technologies.

Biased Technologies

Do a quick Google search for the word “handsome.” About 90 percent of what will pop up will be white males. The results are like this because when a white man thinks of the word handsome, they only think of handsome in their worldview — which happens to be predominantly white people.

This same scenario can be run, on a much larger scale, when algorithms are executed, and this is when racial assumptions can be made through technology. Black people have been mistaken for gorillas in facial recognition technologies, and cameras have mistaken Asian people’s eyes as shut in camera software. Again, this is not likely active racism, rather it’s that the people who program these technologies simply just aren’t taking minorities into account for all-encompassing data for everyone.

Until the tech industry learns to build more diverse companies, this will continue to happen and may get worse. Different minds and perspectives are needed to develop more comprehensive algorithms. Without the inclusion of minorities, a white male perspective is all the data an algorithm will receive.

Ben Dickson goes more in-depth in a great article about algorithmic bias. We can combat algorithmic prejudices by inviting a more inclusive atmosphere in the tech industry. To execute machine and deep learning (predictive algorithms) accurately, it makes sense that the more data points you enter, the better the algorithm will be. So, for the sake of the best programming, the tech industry must include women and minorities — not only because it is the ethical thing to do but also to create the best programs possible.

Desmond Rhodes is a freelance writer, specializing in Big Data, Marketing, and Social Media.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.