Posted 2/22/2024, 7:17:00 AM
Google Apologizes for Racial Bias in New AI Model Image Results
- Google issued an apology after its new AI model Gemini generated racially biased image results in response to user queries
- The controversy has drawn criticism mainly from conservative voices perceiving Google as politically left-leaning
- Users highlighted the difficulty in generating images of white individuals using Gemini, with requests often yielding non-white results
- Some conservative voices framed the outcomes as Google deliberately trying to diminish white representation, with one using coded antisemitic language
- Concerns remain about biased AI reinforcing existing stereotypes and contributing to systemic discrimination