Technology is supposed to be blind. Just a never ending stream of ones and zeros flowing across a universe of wires and devices. Don’t believe it!
Again and again we witness the discriminatory use of technology against people. But lets be clear; its not the technology. Its people behind the technology. Where humans are involved there will be bias. Period.
HUD Hits Facebook
HUD has officially accused Facebook of violating the Fair Housing Act through its ad-targeting tools. The tool allows sellers to limit listings based on categories like race, sex and nation of origin.
HUD has focused its gaze on extremely specific ad-targeting features offered by these companies seeking to determine if they facilitate housing discrimination. Hyper specific advertising allows the owner of an apartment complex to purposefully exclude certain groups, such as people of color, from seeing new listings in violation of the law.
But this type of discrimination is not limited to housing. It has been found in job listings, financial services and even online pricing.
We have to ask if this is a deliberate act by Google, Facebook and Twitter? Probably not. But then again we have to look at these companies as “un-indicted co-conspirators.”
Often when a company places ads on these platforms it is highly automated. Very little actual human interaction takes place. Its all about mouse clicks. But at the same time these companies ignore the potential for bad actors to use the system in unintended ways. Humans program computers and write code and algorithms that allow their misuse. Then look the other way until something like the HUD actions hits them in the head.
The Algorithmic Accountability Act
Now the problem has become so bad that it has caught the attention of lawmakers. A bill introduced b Sen. Ron Wyden and Cory Booker and Rep. Yvette D. Clarke is named The Algorithmic Accountability Act. Although not yet law the bill would require companies to assess the risks of their algorithms on consumers and ensure they are implemented in ways that don’t result in “inaccurate, unfair, biased or discriminatory decisions impacting Americans.”
The bill would require companies to create assessments of their algorithmic systems including “… detailed descriptions of the automated decision system, its design, its training, data, and its purpose.”
Among the requirements are assessments that include the system’s benefits and costs and details on data minimization practices. The company would have to answer questions like, how long data and decisions are stored, what information is available to consumers, if consumers can appeal any of the system’s decisions, and who ultimately these decisions are sent to. They would also include the potential risks of the system, and how they plan to minimize them.
What price are you paying?
Algorhitms are responsible for serious discrimination. Even when you shop online. Woud you be surprised to learn that algorithms are probably making decisions about how much you pay online? According to the Wall Street Journal algorithms are setting the price for all sort of goods including airline tickets. You maybe paying more based on your online activity or other data known about you. Including your known income, credit score, where you live and of course racial and ethnic profiles.
There have been examples where a person’s online profile reveals that a person is economically well positioned, i.e rich. This person may pay a higher price for a product. Especially designer brands or items with limited availability. On the other hand a brand will use algorithms to limit advertising to certain audiences or groups as they are thought to “diminish the brand.” Algorithms cut both ways!
Breaking It Down
All too often people talk about tech diversity. And all too often it is dismissed or minimalized. The result is that you end up with one-sided leadership that can’t seem to grasp the value of diversity in its product offering or, in this case, its programming.
Now we see these same companies dishing out fat checks to lawyers to explain why they did or didn’t do something. When the government gets involved its only then that they see the light. Black people called that a “Come to Jesus moment.”
Diversity means that you can avoid these situations because you get more than one way to look at these things before it all goes wrong. Is tech diversity the only asnwer? Hell no! But injecting some color into the situation can’t hurt. Especially at the executive level.
But there is one final item I need to adress. There are a lot of wealthy people in the world. And they got that way by not paying too much for anything. Even the comfortable middle class should be annoyed by the fact that an algorithm sees them as a good target to milk a few extra dollars out of. See, its really not all about race. Think about that.
Are you starting to get the picture now?