Mr. Sabarna Choudhury

Advancements of AI in Semiconductor Industry

Abstract:

The semiconductor industry, valued at billions across the world, is undergoing a transformative shift driven by the integration of artificial intelligence (AI) across the VLSI (Very Large Scale Integration) design and manufacturing lifecycle. As chip complexity escalates with billions of transistors and shrinking process nodes, AI-powered methodologies are delivering substantial efficiency gains-up to 40% in design and 60% in product development cycles-by automating complex design, verification, and optimization tasks.

AI-driven tools now automate logic synthesis, gate-level optimization, and early issue detection, enabling faster and more reliable RTL (Register-Transfer Level) design. Machine learning models enhance test pattern generation, achieving 95% fault detection accuracy and halving testing times, while AI-assisted physical design tools optimize placement, routing, and power consumption, reducing design iterations by as much as 65%. Advanced neural networks and predictive analytics further drive down power usage and improve thermal management, with real-world platforms demonstrating up to 25% reduction in chip power consumption and 98% accuracy in hotspot prediction.

In timing analysis, AI enables predictive path delay modeling and automated design adjustments, reducing validation cycles by up to 70% for advanced nodes like 5nm. Looking ahead, the industry is moving toward autonomous chip design, where AI systems generate complete, optimized layouts with minimal human intervention and adapt in real time to manufacturing variations. The emergence of specialized AI hardware for VLSI optimization is accelerating this virtuous cycle, with analysts projecting a tripling of the semiconductor innovation rate by 2030. Collectively, these advancements position AI as a cornerstone of next-generation semiconductor innovation, driving unprecedented improvements in speed, reliability, and performance.

Profile:

Sabarna Choudhury is a professional specializing in Very Large-Scale Integration (VLSI), with a profound expertise in Machine Learning and Artificial Intelligence. Holding a master’s degree from the University of Minnesota, USA, he has established a strong academic foundation that complements his extensive industry expertise. Over seven years, Sabarna has contributed to leading technology firms, including Intel, Samsung, and Qualcomm. His dedication to research and development in semiconductor technologies reflects a steadfast commitment to advancing innovation while thoughtfully contributing to transformative solutions in the ever-evolving field of VLSI and AI. His research interests encompass VLSI Physical Implementation, Machine Learning and Artificial Intelligence Classification, and Biomedical Image Processing, reflecting a broad and interdisciplinary approach to advancing technological innovations.