Major systems are increasingly influencing global perspectives through their depiction of the world. These powerful tools, trained on massive datasets, resemble societal values, but also have the potential to amplify existing prejudices. It's imperative that we aim for representative training data and algorithms to ensure that major models cultivat