Links for physical-ai-interpretability