The ongoing digitization of the industrial-scale machines that power and enable human activity is itself a major global transformation. But the real revolution-in efficiencies, in improved and saved lives-will happen as machine learning automation and insights are properly coupled to the complex systems of industrial data. Leveraging a systems view of real-world use cases from aviation to transportation, I contrast the needs and approaches of consumer versus industrial machine learning. Particularly, I focus on three key areas: combining physics-based models to data-driven models, differential privacy and secure ML (including edge-to-cloud strategies), and interpretability of model predictions.