Rachel Berkowitz of IEEE Spectrum reports that TSMC has partnered with startup Avicena to develop microLED-based optical interconnects for AI data centers, aiming to replace traditional copper wiring with a more energy-efficient, cost-effective solution. Avicena’s LightBundle technology uses hundreds of blue microLEDs and imaging fibers—avoiding complex, costly lasers—to transmit high-speed data across short distances. The approach leverages mature components from LED and display industries, enabling faster scaling and lower energy use than current silicon photonics solutions. With prototypes showing superior efficiency, this innovative technology could reshape AI data infrastructure. Berkowitz writes:
In the race to an all-optical AI data center, a major player has now placed a bet on a different horse. Semiconductor manufacturing giant TSMC announced that it will work with Sunnyvale startup Avicena to produce microLED-based interconnects. The technology is a pragmatic twist on replacing electrical connections with optical ones to meet the high needs of communication among an increasing number of GPUs in a low cost, energy efficient way.
Thanks to the computational demands of large language models and their cousins, AI clusters are facing unprecedented requirements regarding amounts of data, bandwidth, latency, and speed. Sooner or later, the copper wires that connect processors and memory within a single AI data center rack will have to be replaced with optics. “There’s a huge push to get optical connections as close to the board as possible,” says Lucas Tsai, a vice president at TSMC. […]
Pezeshki acknowledges that Avicena has a ways to go to build and scale the product. But “the combination of showing great results together with using mature building blocks is winning over converts,” he says.
Read more here.
