Nvidia has committed at least $6.5 billion to companies developing imaging technology in the past three months, as the company rushes to invest in solving one of the biggest challenges in the rollout of artificial intelligence.
Photonics, the use of light to transmit data, is an emerging technology that is considered a more efficient alternative to the current process of transmitting data using electricity. Electronic data transmission consumes more energy – a factor that is increasingly seen as a barrier to the widespread use of AI.
Since early March, Nvidia has announced a $2 billion investment Lumentum, Compatibility again Marvellall who develop photographic technology. The giant said he would invest $500 million in it Corning to develop advanced connectivity solutions, and participated in Optics startup Ayar Labs’ $500 million Series E funding round.
“Photonics represents a way for Nvidia to expand its AI infrastructure without the energy costs that would remain with electricity and copper,” Alvin Nguyen, senior analyst at Forrester, told CNBC.
“By investing in photonics companies, Nvidia ensures that the development of graphics continues and prevents them from hitting the wall of scalability and performance that would happen if they stayed on electricity and copper.”
Solving problems
Photonics can be used in AI infrastructure by using light to transmit data between graphics processing units (GPUs), memory, communication chips, servers and data centers, instead of relying only on copper-based electrical signals.
While copper is the primary communication standard today due to its low cost and high reliability, photonics will dominate the AI infrastructure over time, Brian Colello, senior equity analyst at Morningstar, told CNBC.
“Nvidia’s roadmap for next-generation AI rack-scale solutions will require an increasing number of virtual interfaces to process the increased bandwidth with new models and higher utilization,” he said.
The chip giant is already making some photonics technology available as part of its networking solutions, with the company announcing tools it says will enable AI factories to connect millions of GPUs across sites while significantly reducing power consumption and operating costs.
“If you look upstream, you come to the conclusion that we are starting to scale our silicon photonics technology,” Nvidia CEO Jensen Huang said at GTC in March, pointing to Nvidia’s ethernet network platform used to connect industrial AI and GPU clusters. He also said the company is starting to add photonics to its GPU-to-GPU communication technology.
“Which means the amount of silicon photonics technology we need is much higher than the world has today,” he added. “So we’re working with the supply chain to make sure we can help them build capacity before then.”
Lumentum stock is up 134% since the start of the year, while Coherent is up 96%. Marvell saw its shares rise 122% in 2026 and Corning 111%.
Shares of companies involved in photography have risen over the past year.
Nvidia is one of many AI players that have recently moved into photonics tech.
A chip maker AMD joined Nvidia in the Ayar Labs round, as well as acquiring the first Enosemi in 2025, as well as investing equity in Teramount and Celestial AI. Alphabets again Microsoft Venture Arms backed neye in an $80 million Series C in April.
But deploying graphics technology across AI infrastructure stacks at scale comes with its own challenges.
“The technology is sound, the productivity scale is a difficult problem,” Nick Patience, who leads AI at Futurum Group, told CNBC.
“Manufacturing yield in integrated optical assemblies is still a challenge because the precise alignment of the optical and silicon components is not forgiving, and if something goes wrong in the packaging process, the assembly often cannot be reworked,” he said.
So the transition is underway, but it’s still very early days,” Patience added. “I would expect us to see significant adoption from 2028 onwards.”
