The first demand was for Amazon to stop selling this software to police, law and government agencies. Second, was to set up a human and civil rights inspection into the usage of the software.
The shareholders voted to reject both. Moreover, the proposals are non-binding. This implies that Amazon won’t have to take any sort of responsibility even with a majority vote.
It’s absolutely astonishing that pressing issues regarding human and civil rights required a vote at Amazon. Especially when facial recognition software has a history of unfairly targeting people of color, minorities, and women. The more concerning part is this problem isn’t only limited to Amazon. Almost all big tech company’s software has shown a racial bias or taken part in government surveillance, which is completely racist, misogynist and dangerous.
For instance, earlier this year, Microsoft was accused of working with a Chinese military university on research regarding facial recognition which could be used in the persecution of China’s Muslim minority- the Uighurs. If you haven’t already heard China has been detaining Muslim minority groups in prison-like conditions and forcing them to denounce their religion.
The country has already set up several facial recognition cameras that can track citizens’ whereabouts. Furthermore, the data collected through facial recognition software can be accessed by the police easily.
Although, Microsoft did bar the selling of its facial recognition to law enforcement in reply to the whole debacle. it still doesn’t address the bigger issue. Why are big tech companies selling consumer products to governments who use these to further impede on privacy and human rights? Seems like there’s no ethics when it comes to capitalism.
In recent events, Ousmane Bah, an 18-year-old college student sued Apple for misidentifying him for a thief using facial recognition. This isn’t the first time a person of color has been misidentified as a criminal, and definitely not the first time using this technology. It has been proven that false matches affect people of color at a higher rate, especially women of color and this can be extremely dangerous, even deadly. In fact, 35% of dark-skinned women were mistaken as men by facial software. In contrast to misidentification of light-skinned women was around 7%. Additionally, Transgender drivers are getting kicked off of Uber due to facial recognition not recognizing them because it can’t pinpoint their gender.
Facial recognition will only be effective and accurate on the people it’s trained on which is, primarily, white men. The technology is 99% accurate on this demographic. With the lack of diversity and the lack of inclusion in new upcoming technologies using anything designed for one demographic in law enforcement can deepen the injustice against people of color. Especially, when there is already a pre-existing racial bias in the police force against people of color. Reports show that African-Americans are more likely to be stopped and searched by the police. Arming a police officer with a piece of technology that works against people of color is detrimental.
Tech companies have a large amount of influence over society, economy, and democracy. In fact, more than half of e-commerce goes through Amazon. Furthermore, Amazon has also been under fire for reportedly meeting up with U.S. Immigration and Customs Enforcement (ICE) to market their Facial Recognition software.
It is evident that human bias sneaks into algorithms because they are being created by scientists who have these racial biases and if this continues to go unchecked it will continue to be a detriment to society.
Jacob Snow, a tech and civil liberties lawyer, in his statement explains “Face surveillance will be used to power discriminatory surveillance and policing that targets communities of color, immigrants, and activists. Once unleashed, that damage can’t be undone.”