Ayar Labs joins NVIDIA NVLink Fusion ecosystem - Engineering.com

Ayar Labs has joined NVIDIA's NVLink Fusion ecosystem, marking a strategic integration of its optical interconnect technology with NVIDIA's AI infrastructure platform. The company's co-packaged optics... Read original →

Ayar Labs’ integration into NVIDIA’s NVLink Fusion ecosystem elevates optical interconnects from a niche enhancement to a foundational AI infrastructure layer. This move pressures legacy copper-based SerDes toward obsolescence in high-end GPU clusters, triggering cascading redesigns across packaging, substrates, and thermal systems. Amid tightening U.S. export controls on advanced compute, the partnership deepens NVIDIA’s vertical lock-in—but exposes Taiwan, China and mainland Chinese hyperscalers reliant on U.S.-sourced optical engines to heightened supply chain compliance risks. Rivals like Intel and Broadcom may counter with silicon photonics paired with open standards (e.g., UCIe), while Marvell could accelerate acquisitions in optical modules. Within 18 months, CPO will shift from optional to mandatory in AI fabric designs, driving >30% reduction in bandwidth-per-watt costs and compelling OSAT leaders like TSMC and ASE to fast-track co-packaged optics capacity—or risk losing next-gen AI factory contracts.