Description
When a turbine bearing starts overheating, you can’t wait for data to make a round trip to the cloud and back. That split-second decision needs to happen on the spot – and that’s exactly what edge AI is designed for.
In this episode of Getting Real With AI, host Michelle Kiener is joined by Prof. Melika Payvand from the Institute of Neuroinformatics at ETH and the University of Zurich, and Ankush Gulati, ABB’s Global Program Lead for Energy Efficiency, joining from Singapore. Together they explore what edge AI actually is, how it differs from cloud-based approaches, and why the human brain turns out to be a surprisingly useful model for understanding it.
Topics covered include real-time anomaly detection, data sovereignty, predictive maintenance, neuromorphic computing and resistive memory, and what bio-inspired hardware design could mean for the future of industrial AI.
