Biases in large image-text AI model favor wealthier, Western perspectives: Study

In a study evaluating the bias in OpenAI's CLIP, a model that pairs text and images and operates behind the scenes in the popular DALL-E image generator, University of Michigan researchers found that CLIP performs poorly on images that portray low-income and non-Western lifestyles.

ai model bias biases clip computer sciences dall dall-e found generator image image generator images income low openai perspectives popular researchers study text university university of michigan