In recent years, there has been much talk surrounding the implications of artificial intelligence (AI). As AI becomes more and more pervasive in everyday life, from banks to police forces, there is a growing body of evidence that suggests that AI is not immune to systemic racism.
The main argument for this comes from a study in 2018 demonstrating that AI does not have the same algorithm for everyone, making it unfortunately prone to the same racial biases that are present in society. The study found that the facial recognition algorithms used in many AI technologies have a racial bias, only accruing accurate recognition results when the target’s skin color is light. This means that AI technologies are more likely to make mistakes when identifying people of color, leading to possible outcomes such as wrongful arrests and perceptions of criminal behavior due to race.
The implications of AI bias don’t just stop at facial recognition programs – it’s pervasive across the range of AI technology being developed. Banking algorithms, for example, are increasingly being used to determine loan eligibility or creditworthiness. But these same algorithms are found to be more likely to reject loans to people of color, further entrenching systems of oppression.
On top of this, the development of AI technology is only increasing the divide between different social classes. The algorithms used to create AI technology are only as good as the people who create them, and this means that those at the top can design algorithms to favor themselves and maintain the status quo.
Although the implications of AI bias are concerning, it is important to note that these problems can be addressed. There are a myriad of different legal and regulatory tools available to governments to prevent the perpetuation of systemic racism through AI technologies. Moreover, companies and organizations building AI must take steps to ensure their models are not serving to perpetuate existing racial hierarchies.
In this age of widespread technology, it is evident that the potential for AI to entrench systemic racism is an all too real problem that must not be overlooked. As Nations and organizations continue to develop AI, they must be conscious of the potential for AI to do harm, and must be held accountable for tackling the issue. Only then can we ensure that AI technology is used in a fair and equitable way.