Photonics is emerging as a new key link in the AI infrastructure race beyond GPUs

📌 The AI infrastructure story is expanding beyond traditional chips, as investors pay closer attention to photonics and optical networking. Bloomberg Technology on June 18 highlighted that data center connectivity could become the next major focus as AI workloads continue to scale.
🔎 The core issue is the “copper wall,” referring to the limits of copper-based connections as AI clusters grow larger. When tens of thousands of GPUs need to move data continuously, the challenge is no longer only computing power, but also bandwidth, latency, energy use and heat.
⚙️ This is why optical networking and silicon photonics are gaining more attention from the market. Transmitting data through light can help improve speed and energy efficiency, especially inside large-scale AI data centers.
📈 This narrative carries more weight as Nvidia has made major investments in optics-related companies such as Lumentum and Coherent. It suggests that the next AI bottleneck may not only be GPUs, but also the ability to connect the entire system efficiently.
#AIInfrastructure