Google Executive Apologizes for AI’s Racial Bias Issue

According to a report by the Associated Press on February 24th, Google’s AI model Gemini has been found to have “anti-white” bias. Google’s official statement and its executives have recently apologized for this issue.

Critics have pointed out that Gemini portrays specific white figures as people of color. This phenomenon is seen as an overcorrection of the long-standing racial bias issue in the field of AI.

For example, a user on social media platform X provided text describing the “Founding Fathers of America,” but the images generated by Gemini depicted all individuals as Black, which does not align with historical facts.

Image Source: Social Media Platform X

Google released a statement on this issue stating, “We acknowledge that Gemini has provided inaccurate information in some historical image generations. We are actively working to rectify this situation. While Gemini is capable of generating a wide range of images, which is generally beneficial, it has missed the mark in this case.”

Image Source: Social Media Platform X

Furthermore, Prabakar Ragavan, the senior vice president responsible for Google’s search engine and other businesses, stated, “Clearly, this feature did not meet the standards. Some images generated are inaccurate and even offensive. We appreciate the feedback from users and apologize for the malfunction of this feature.”

Source: DCCN Economy News Aggregated