Google Apologizes for Gemini AI’s Racial Bias

Google Apologizes for Gemini AI's Racial Bias

Google has apologized for Gemini's inaccuracies in depictions of historical images. Google's Gemini has been criticized for depicting images of specific white figures like the US Founding Fathers, and Nazi-era German soldiers as people of color. Gemini has received a lot of backlash on X for generating inaccurate images of historical facts.

On the inaccurate depiction of historical images, Google stated that they are working to improve these kinds of inaccurate depictions immediately. Additionally, Google also stated that Gemini generates a wide variety of images of people and this is a good thing as people from all over have been using it but it has been missing the mark on the historical images.

Earlier this month, Google began offering the image generation feature in the AI-powered tool Gemini. It is not the first time that AI has stumbled into questions over racial bias and diversity. Even earlier about a decade ago Google had to apologize when the photo app conveyed a black couple as gorillas.

According to a report, Google does not generate those images which it felt were errors but it is possible that attempted to boost diversity by addressing a chronic lack of it in generative AI. The image generators are trained to generate the best picture and text captions, which means that they can sometimes amplify stereotypes.

Some of the people who criticized as well as defended Gemini wrote on X that it's good to depict diversity in some cases and criticized the fact that Gemini is inconsistent.

Another X user wrote that it's quite embarrassing to get Google Gemini to acknowledge that white people exist.

Currently, Gemini has been refusing to create images of Vikings or Nazi-era German soldiers.