According to the researchers of the University of Cambridge, the depiction of artificial intelligence in culture as white can carry a lot of consequences, which includes the erasure of people who are not white.
Researchers say that the depiction of artificial intelligence (AI) in white can impress people trying to enter the field of AI as well as managers making hiring decisions. This also causes serious consequences. It’s been said that whiteness is not only seen as an AI assistant with a stereotypically white voice, is also seen as the absence of color, the treatment of white by default.
White depictions of AI were usual in Google search results for “Artificial Intelligence Robot”. This depiction of artificial intelligence (AI) as default white can distort the perception towards risk and opportunities related to predictive machines thereby creates questions in and around middle-class white people.
An algorithmic bias has been recorded in a range of technology in recent years, from both automatic speech detection systems and popular language models related to health care, lending, housing, and facial recognition. Bias is found in areas like occupation, religion, sexual identity as well as in race and gender.
According to a study conducting in the year 2018 shows that majority of the participants attributed a racial identity to a robot based on the machine’s exterior color. But then, a research paper of 2018 shows that when the participants in a study involving Black, East Asian, and white robots were likely to deprive the language when interacting with Black and East Asian robots.
The debt of “The Whiteness of AI” is made after the release of a paper in June by a Ph.D. student about “how to combat anti-Blackness in the AI community”. An anti-approach was called by the researchers to fight against AI’s problem with whiteness. In an AI workshop at ICML Queer, DeepMind research scientist and AI paper co-author Shakir Mohamed suggested queering machine learning as a way that could be used to bring more equitable forms of AI.
Researchers say whiteness can even reshape the perception of what a robot uprising may look like, incorporating attributes like power and intelligence. They also say that an additional investigation is required to understand the impact of whiteness in the field of Artificial Intelligence (AI).
