6 insights from the GSA Women’s Leadership Initiative: AI semiconductor industry trends turn the chip race into a systems race
For decades, the semiconductor industry competed on one thing: the individual chip. One of the biggest AI semiconductor industry trends now under way is the shift toward treating chips as components of much larger systems. Performance, cost and reliability are now determined across the package, the rack and the supply chain, not inside a single […] The post 6 insights from the GSA Women’s…
The GSA Women’s Leadership Initiative showcased six key insights into the evolving trends in the AI semiconductor industry. These trends are shifting the focus from individual chips to systems, with performance, cost, and reliability now determined across multiple layers such as the package, rack, and supply chain. Optical interconnects are becoming increasingly important to address the limitations of data transfer between chips, memory, and racks, with copper and optics being used side by side.
Qualcomm's High Bandwidth Compute is a low-power technology designed to attach compute to memory, optimizing for tokens per watt. This trend highlights the need for early involvement of product engineering in design-for-test, manufacturing, and reliability planning. Intel's EMIB bridge technology enables scaling packages from 3x to over 10x in form factors, while outsourcing assembly and test has transformed into system integration.
Advanced packaging, optical interconnects, and AI-driven yield learning are key areas of focus. Physical AI extends beyond humanoid robots, encompassing any autonomous system that senses its surroundings, reasons, and acts in the physical world. The market is at its initial stage, and collaboration among companies will be crucial to develop architecture, security, and safety standards.
Firmware is becoming the crucial layer that ties various components together in AI systems, with security extending beyond secure boot to continuous trust during system operation. Sourcing strategies are evolving to include strategic partnerships with suppliers, who receive product roadmaps early to align with AI-driven demand changes. Leading-edge logic capacity is a major constraint, followed by packaging, specialty materials, and critical equipment.
To address these challenges, companies are adopting dual sourcing, spreading manufacturing footprints across regions, and integrating AI into electronic design automation flows to reduce development cycles and improve verification closure.
Written by urgent.news from SiliconANGLE's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
