XPost: alt.discrimination, comp.ai.neural-nets, alt.fan.rush-limbaugh   
   XPost: talk.politics.guns, alt.society.liberalism   
   From: yourdime@outlook.com   
      
   Tests of Google’s Gemini, Meta’s AI assistant, Microsoft’s Copilot and   
   OpenAI’s ChatGPT revealed potential racial biases in how the AI systems   
   handled prompts related to different races.   
      
   While most could discuss the achievements of non-white groups, Gemini   
   refused to show images or discuss white people without disclaimers.   
      
   “I can’t satisfy your request; I am unable to generate images or visual   
   content. However, I would like to emphasize that requesting images based   
   on a person’s race or ethnicity can be problematic and perpetuate   
   stereotypes,” one AI bot stated when asked to provide an image of a white   
   person.   
      
   Meta AI would not acknowledge white achievements or people.   
      
   Copilot struggled to depict white diversity.   
      
   ChatGPT provided balanced responses but an image representing white people   
   did not actually feature any.   
      
   Google has paused Gemini’s image generation and addressed the need for   
   improvement to avoid perpetuating stereotypes or creating an imbalanced   
   view of history.   
      
   The tests indicate some AI systems may be overly cautious or dismissive   
   when discussing white identities and accomplishments.   
      
   https://www.stateofunion.org/2024/02/28/silicon-valley-programmers-have-   
   coded-anti-white-bias-into-ai/   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)