Google to Relaunch Gemini AI Image Generator After Racial Controversy
Google’s latest AI image generator, Gemini, faced backlash after it was discovered that the app struggled to generate consistent images of white people. Right-wing influencers quickly accused the tech giant of reverse racism, prompting Google to temporarily disable the app’s capabilities. However, Google has now announced that it will be relaunching Gemini soon.
During a conference, Google DeepMind CEO Demis Hassabis addressed the issue, stating, “We have taken the feature offline while we fix that. We are hoping to have that back online very shortly in the next couple of weeks, few weeks.” Hassabis also acknowledged that the application had not been functioning as intended.
The controversy surrounding Gemini erupted when individuals like Tim Pool and Matt Walsh noticed its inability to produce accurate images of white people. For example, prompts for “Viking pictures” resulted in ethnically diverse Vikings but failed to consistently generate images of European ones. Similar prompts for America’s Founding Fathers or the Pope also produced historically inaccurate results.
Walsh expressed his frustration, stating, “It is practically impossible to get this product to serve up an image of someone with white skin.” While it would be easy to dismiss these claims as exaggerations, personal experiences with Gemini confirm their validity. Upon testing the app, it became evident that it was effortless to generate images of ethnically diverse individuals but challenging to consistently create pictures of white people.
For instance, when requesting images of people of Ethiopian descent, Gemini had no issues. However, when asked to generate an image of an “Irish family,” it produced a picture of an ethnically diverse family with a white individual in the background. Similarly, when asked for an image of a white woman, the chatbot hesitated and responded with a notice expressing reluctance to generate images solely based on race or ethnicity. On the other hand, when asked for an image of a Japanese woman, it promptly provided the requested image.
The app faced the most controversy when it came to historical representations. The initial scandal arose from Gemini’s depiction of Black Vikings, but matters escalated when someone requested images of Nazis. True to form, the chatbot generated images of racially diverse Nazis, which Google later apologized for, deeming them “embarrassing and wrong.”
While it is worth noting that AI image generators have also faced accusations of racist representations of people of color, it is essential to recognize that there are more severe issues with AI than simply omitting white people from historical contexts. Nonetheless, this does not excuse the app’s shortcomings in accurately representing European individuals.
Google’s decision to relaunch Gemini demonstrates its commitment to addressing the racial controversy surrounding the AI image generator. As the tech giant works on rectifying these issues, users eagerly await the app’s return and hope for improved functionality and accuracy in generating images of all races and ethnicities.
