Artificial Intelligence (AI) has produced a decade of promising results in power electronics, yet relatively few are deployed where decisions carry significant economic, reliability, or safety consequences. This talk examines what it takes for AI to become a trusted part of engineering practice. It presents advances from academia and industry, including examples from the speaker's team in modeling, control, reliability testing, and condition monitoring. Four foundations are identified for translating promising results into trusted practice: the right data rather than simply more data, quantified uncertainty and repeatable performance, consistency with physical principles, and an auditable pathway to validation and qualification. The talk concludes by examining generative and agentic AI, from agent-driven design and simulation to self-driving laboratories. It argues that while these technologies may reshape how power electronic systems are designed, tested, and operated, their adoption will rest on the same four foundations.

Huai Wang is a Professor with the Department of Energy (AAU Energy), Aalborg University, Denmark. He leads the Reliability of Power Electronic Converters (ReliaPEC) research group and chairs the Mission on Digital Transformation and AI at AAU Energy. His research focuses on efficient, reliable, and cognitive power electronic components and converter systems, spanning reliability engineering, condition monitoring, artificial intelligence, and digitalization for power electronics. Since joining Aalborg University in 2012, he has established five industrial and PhD courses attended by more than 1,000 PhD students and industry engineers and has delivered over 35 international conference tutorials. He received the B.E. degree from Huazhong University of Science and Technology in 2007 and the Ph.D. degree from City University of Hong Kong in 2012. He was elected a member of the Danish Academy of Technical Sciences in 2023.