Google Apologizes for AI Chatbot Generating Inaccurate, Diverse Images of Historical Figures
-
Google apologized for its AI chatbot Gemini generating inaccurate, racially diverse images of historical figures like the Founding Fathers and Pope.
-
Head of Google's AI division said they are working to fix the historical inaccuracies immediately and take representation and bias seriously.
-
Inaccurate images showed white historical figures as Black, Native American, or Asian, sparking backlash from some conservative users.
-
Google plans to tune the Gemini model to account for more nuanced historical context.
-
This is not the first diversity-related issue for Google, having previously apologized for its photo app labeling a black couple as "gorillas" nearly 10 years ago.