There were 2 sessions on fairness at SoCraCan. One of them was on how to make the recruiting process fairer, and the other one was about fair compensation. Both on them build upon the idea that what we currently have as a system is flawed and creates unfairness, be that because of human bias, subjectivity or any other uncontrollable variable that may impact the result. In a deeper sense, it asks a question:
Are we really capable on fairness? And if we are, what do we need to do to achieve it?
I’m not going to answer that question, mostly because I don’t know if there’s even an answer for it. Nevertheless, since this is all about words, here are mine.
We are not the code we write- But the code we write, carries a part of us with it, just like children would carry around the DNA of their parents, and in most cases, some behavioural aspects as well. Assuming that we can be entirely objective when writing code is ludicrous, because no human activity that cannot be represented with an infallible formula can be entirely objective. Everything we do is filled by our context. Sometimes it’s out in the open staring right at us, and sometimes it hides out in the way our algorithms taught an AI how to recognize human faces. But it’s there. It will always be there. So, any solution to the idea of fairness based on code or software, will be flawed because we are flawed. So, let’s not even go there.
I believe fairness can only be achieved when the right variables are removed, and the right ones are taken into account. The variables worth removing should be anything that a human being did not choose, and cannot control. Gender, Race and Nationality are all variables that can fall into this category. So, let’s start by trying very very very hard to remove them from the idea of fairness. My argument behind that, is that real fairness can only be achieved once every possible uncontrollable aspect has the same value. And if the value we give it is 0, then that’s the same value, isn’t it? Of course, this only works when the scales aren’t heavily tilted in any side, but that’s a discussion for another time.
Next step is taking the right variables into account. This is very highly dependent on whatever it is we’re trying to be fair about, so let’s use the compensation example. What are variables we should include in the conversation of fair compensation? Here’s what I believe should be included and is often ignored:
Morality ➡ A business’ goal is to provide benefit to those involved in it. In some cases, that means the entire human population, and in others a small town or even just a neighbourhood. The size doesn’t really matter. What does matter, is that in order to be a morally responsible business, the first preoccupation must be to not actively create pain, but strive to build benefit by creating value. When it comes to fair compensation, this can mean a couple things:
This is, of course a very simplistic model, that will most probably not work in any way. But it’s worth keeping these variables in mind if you’re in a position of power. And as I know you already noticed, everything in here comes again to the idea that people are more important than anything else. For society and for a business. Even one that makes money with computers. I really do think that if we’re ever capable to put people first, we can accomplish most things. And achieve true fairness. In the meantime, we can still try and make some steps in the right direction.