AI processing of tomorrow needs new hardware. The world needs faster chips to grow its AI capability. We want to place a supercomputer in every person’s hand. To do this, we need to re-think our approach to chip architecture and Photonics is the answer.

Salience Labs is a team of physicists, engineers, and coders building a new type of chip. One that uses light to execute operations. We have designed a massively parallel, ultra-high-throughput tensor processing chip. It uses photonics to enable significant improvements in processing performance. We package it with electronics to create a high-speed, multi-chip processor that accelerates exponential advances in AI. Our technology is based on decades of research from the University of Oxford, UK, and Münster University, Germany. The speed of AI computation doubles every 3.4 months, outpacing what standard semiconductor technologies have left to offer. To accelerate exponential advances in AI across industries, a new paradigm for computing – one that is both faster and highly application-specific - is now required.