The debate surrounding the potential biases of artificial intelligence has become a growing concern in recent years. People fear that AI systems could perpetuate and even amplify existing biases, leading to unfair treatment and the perpetuation of inequalities. This fear is understandable, as we’ve seen numerous examples of AI systems exhibiting biased behavior. However, it’s important to recognize that humans, by our very nature, are flawed and biased creatures. Why, then, do we fear AI bias so much when it’s clear that humans themselves are far more biased?
In this opinion piece, I will explore the nature of human bias, the potential risks of AI bias, and what can be done to address both. I will also discuss the paradox of human fears regarding AI bias and how understanding this contradiction can lead us to seek better solutions for both human and AI biases.
Human Bias: The Inevitable Imperfection
A quote by the famous philosopher Immanuel Kant comes to mind:
Out of the crooked timber of humanity, no straight thing was ever made.
This statement captures the essence of human nature, which is inherently flawed and filled with biases. These biases manifest in various forms, such as cognitive biases, prejudice, and stereotypes, and they are deeply ingrained in our social, cultural, and individual experiences.
The field of psychology has produced extensive research on human biases, revealing just how pervasive they are in our decision-making processes. For instance, the confirmation bias, which is the tendency to search for, interpret, and recall information in a way that confirms one’s preexisting beliefs, has been demonstrated in numerous studies (Nickerson, 1998). Similarly, the availability heuristic, which involves making decisions based on the most readily available information rather than a thorough analysis, has been shown to lead to systematic errors in judgment (Tversky & Kahneman, 1973).
It’s clear that human bias is a formidable challenge to overcome. But how does this compare to the risks of AI bias?