As we’ve been doing research with this initiative, we’ve been thinking a lot about what it means to have religious plurality and spiritual plurality in the representation of information. For example, a lot of religious people are concerned by how these models may output things that are not theologically correct or are even offensive or harmful. I’ve interviewed someone who’s a survivor of image-based abuse, a Muslim woman, talking about how those case studies are not tested enough. There’s so much nuance, case studies, heuristics, and religion/spirituality are a moving force. What do you think about alignment and representation of these concepts in outputs from these models?