Nvidia is investing billions into this emerging technology that could change the AI industry

Nvidia has committed at least $6.5 billion to photonics technology over the past three months, targeting a key bottleneck in AI deployment. Photonics, which uses light instead of electricity to transmit data, promises greater efficiency and lower energy consumption for AI infrastructure. The investments include $2 billion in Lumentum, Coherent, and Marvell, as well as $500 million each in Corning and Ayar Labs, a startup participating in a $500 million funding round. Nvidia’s CEO, Jensen Huang, highlighted the need for silicon photonics to scale AI systems, noting current global capacity falls short of demand. Photonics could replace copper-based electrical signals in AI data centers, GPUs, and networking chips, reducing energy costs and improving performance. While copper remains dominant due to cost and reliability, photonics is expected to grow as AI demands increase. Analysts like Brian Colello of Morningstar note Nvidia’s roadmap requires optical connectivity to handle rising bandwidth needs from advanced AI models. Nvidia has already integrated photonics into its networking solutions, enabling AI factories to connect millions of GPUs with lower energy use. Huang stated the company is expanding silicon photonics capacity in advance of future needs, collaborating with suppliers to meet demand. Stocks of photonics firms have surged this year: Lumentum is up 134%, Coherent 96%, Marvell 122%, and Corning 111%. Other tech giants are also investing in photonics. AMD joined Nvidia in backing Ayar Labs, while Alphabet and Microsoft’s venture arms supported nEye in an $80 million Series C round. However, scaling photonics production remains a challenge, with experts like Nick Patience of the Futurum Group acknowledging that while the technology is viable, mass adoption requires overcoming manufacturing hurdles.