Nvidia to Invest $4B in Companies to Scale AI Infrastructure

Nvidia will be investing $4 billion in two U.S. photonics companies, as it continues to support buildout of AI data center infrastructure.

The chip vendor said it will be channeling $2 billion into Lumentum and $2 billion into Coherent, both of which develop optical technologies that use light, rather than electrical signals, to connect AI chips.

The tech is used for high-bandwidth applications such as sensing and data transfer as well as powering next-generation AI infrastructure.

“Optical interconnect technology and package integration are critical for the continued scaling of AI factories,” Nvidia said in a statement on the deal. “[They improve] the energy efficiency and resiliency of large-scale AI networks.”

As part of the multi-year agreements, Nvidia will support the companies’ technology development and help expand their U.S.-based manufacturing capacity. 

Both deals also include a multi-billion dollar purchase commitment for the companies’ laser and optical components.

Related:Debate Rages Over AI Bubble vs. Boom

Lumentum CEO Michael Hurlston said in the press release that the company would be investing in a new fabrication facility to “increase capacity and accelerate innovation,” while Coherent said it would use the capital to scale manufacturing and support the global buildout of next-generation AI data centers.

“Computing has fundamentally changed. In the age of AI, software runs on intelligence with tokens generated in real time by AI factories for every interaction and every context,” Jensen Huang, founder and CEO of Nvidia, said in a press release.

“This strategic relationship underscores Coherent’s role as a key enabler of next-generation AI data center infrastructure,” Jim Anderson, CEO of Coherent, said in the press release. “We are proud to expand our 20-year relationship with Nvidia … to help them build the AI data centers of the future.”

The announcement comes amid intensifying competition to secure AI infrastructure capacity. Demand for advanced chips and supporting technologies continues to surge as enterprises across industries scale AI workloads.

Just last week, one of Nvidia’s top customers, Meta, signed a $100 billion deal with AMD in which it agreed to buy up to 6 gigawatts of AI computing power from the chipmaker.