Capital aims to secure optical components for AI expansion
Nvidia is deploying a combined $4 billion into two U.S. companies that build photonics and optics technologies, linking the investments to long term supply arrangements as demand for AI infrastructure accelerates. The chipmaker said Monday it will invest $2 billion in Lumentum and $2 billion in Coherent, a move aimed at reinforcing research pipelines and shoring up supply chains for critical optical parts.
Optics and photonics systems generate or transmit light and are used in areas such as sensing and data transfer. As AI computing scales, these components become increasingly central to moving data efficiently across networks and within large data center environments. Nvidia’s decision to pair equity investments with purchase commitments signals an effort to secure access to hardware that can become constrained when demand rises rapidly.
Investors welcomed the announcement. Lumentum shares gained about 8% and Coherent climbed roughly 13% following the news, reflecting expectations that the deals may bring durable orders and capacity visibility.
Lumentum agreement includes purchase commitments and capacity rights
Lumentum, based in the United States, develops optical and photonic technologies used in the networks and infrastructure that support AI, cloud computing, and next generation communications. Nvidia said its relationship with Lumentum will be structured as a multi year strategic agreement rather than a one time investment.
The agreement includes a multi billion dollar purchase commitment from Nvidia and future capacity rights for advanced laser components. Such terms are designed to give Nvidia priority access to specialized parts while providing Lumentum with clearer demand signals that can support its own production planning and research investment.
Jensen Huang, Nvidia’s founder and chief executive, framed the partnership as part of the company’s broader infrastructure strategy. He said, “Together with Lumentum, NVIDIA is advancing the world’s most sophisticated silicon photonics to build the next generation of gigawatt-scale AI factories.”
Coherent partnership targets next generation silicon photonics
Coherent, also U.S. based, builds photonics technology that uses photons to create components and systems for high performance optical applications. Nvidia said it will collaborate with Coherent on developing next generation silicon photonics for AI infrastructure, positioning the partnership as a technology and supply effort rather than a passive financial stake.
The arrangement includes a multi billion dollar purchase commitment and future access and capacity rights for advanced laser and optical networking products. The structure mirrors the Lumentum deal and indicates Nvidia is pursuing multiple supply channels for similar categories of components to reduce dependence on a single supplier.
Huang said Nvidia will work with Coherent on next generation silicon photonics, emphasizing the role of optical interconnects and laser based hardware in enabling higher performance systems as computing clusters grow in size.
Photonics moves into the center of the AI hardware roadmap
Nvidia’s announcement highlights how the AI buildout is increasingly constrained by connectivity and data movement, not only by computing chips. Optical technologies support faster data transfer and can be critical in large scale deployments where latency and throughput shape overall system performance.
By combining capital with long term procurement commitments and capacity rights, Nvidia is seeking to secure both innovation and availability. For the suppliers, the agreements provide funding alongside multi year commercial support, potentially accelerating development and production of advanced components used in optical networking and AI infrastructure.
The company’s emphasis on gigawatt scale AI factories suggests it expects continued growth in data center buildouts that require more sophisticated networking hardware, creating a rationale for deeper partnerships in photonics as part of the broader AI supply chain.
