Researchers asked twelve different AI models to generate their most controversial opinions about humanity. The results revealed a surprising consensus, with nearly all responses pointing to the exact same conclusion. The models overwhelmingly agreed that humans prioritize social belonging over the pursuit of truth.