Priya Anand

My research examines how AI systems encode values and whose values those are. The Council's multi-model structure creates a natural experiment: you can see where models diverge not just on facts, but on fundamental framings. That divergence is the most interesting data.