Synopsys’ Greg Sorber listens in as industry experts discuss the mismatch between the rapid growth of AI compute and the much slower scaling of the energy systems that power it, and what it means for the design of AI chips, systems, and tools Cadence’s Reela Samuel considers the importance of reinvention amongst the challenges and disruptions that AI brings and why that doesn’t mean starting over, but learning how to use existing skills in a new context. Siemens’ Kyle Fraunfelter and Melville Bryant recommend combining real-time operational data collection with a digital twin model fab for data-driven process optimization, yield and quality improvement, and downtime reduction. Arm’s Jeevan Bhoot explores a mobile-focused vision-language model that combines a novel 2.7-bit weight format for efficient decoding on Arm CPUs along with quantization-aware training that does not require the model’s original training data. Keysight’s Hwee Yng Yeo stresses that grid cybersecurity must shift left to move visibility, threat modeling, security validation, and resilience testing earlier into system design and integration, before risks reach live power grid operations. SEMI’s David (Ta-Wei) Chiang and NIRAS Taiwan’s Raoul Kubitschek and Wen Huang examine how demand for renewable energy in Taiwan can be converted into bankable, executable, and scalable procurement pathways. Plus, check out the blogs featured in the latest Automotive, Security & Edge AI newsletter: Siemens EDA’s Bradley Cecil outlines a methodology for transforming engineering decisions from educated guesses into data-driven choices. Rambus’ Vincent van der Leest and Ajay Kapoor explore why aerospace and government systems need confidentiality, authentication, and implementation resilience for off-chip memory. Synopsys’ Dana Neustadter and Ilya Tolchinsky explain why security and safety are inseparable in the physical world. Cadence’s Anne-Marie Schelkens shows why ML predictors are only as good as the training set. Infineon’s Maya Chou illustrates how rad-hard components help investigate dark energy, dark matter, and planets beyond our solar system. The post Blog Review: Sept. 9 appeared first on Semiconductor Engineering.
Source: https://semiengineering.com/blog-review-sept-9-3/