The problem with waiting for quantum computing to bring in the next wave of AI is that it will likely arrive much slower than humans would like. The next best option is to somehow increase the speed of existing computers, but there is now an important added value: energy-efficient systems that mean we don’t burn the planet while keeping the AI singularity alive. call.
Meanwhile, the speed of AI computation is doubling every 3 or 4 months, meaning standard semiconductor technologies are struggling to keep up. Several companies are now working on ‘photonics processing’ that introduces light into the semiconductor region, which, for obvious ‘speed of light’ reasons, literally accelerates the whole thing considerably.
Salience Labs is an Oxford-based startup that believes it has the answer, combining an ultra-fast multi-chip processor that packages a photonics chip along with standard electronics.
It has now raised a $11.5 million seed round led by Cambridge Innovation Capital and Oxford Science Enterprises. Also participating were Oxford Investment Consultants, former CEO of Dialog Semiconductor Jalal Bagherli, ex-Temasek board member Yew Lin Goh and Arm-backed Deeptech Labs.
Salience is a 2021 spinout of the University of Oxford and the University of Münster, after it came up with the idea of using a wide bandwidth of light to perform operations to deliver what it calls “huge parallel processing performance.” within a given wealth envelope”. † The company says the technology is highly scalable, stacking up to 64 vectors in a beam of light.
Vaysh Kewada, CEO and co-founder of Salience Labs, told me, “This technology will mean we can do a lot more calculations for the same power requirement — which means fundamentally more efficient AI systems.”
She thinks the world needs increasingly faster chips to grow AI capabilities, but the semiconductor industry can’t keep up with this demand. “We solve this with our patented ‘on-memory compute’ architecture that combines the ultra-high speed of photonics with the flexibility of electronics and the manufacturability of CMOS. This will usher in a new era of processing, where supercompute AI will become ubiquitous,” she said.
Ian Lane, Partner, Cambridge Innovation Capital added: “Salience Labs brings together deep domain expertise in photonics, electronics and CMOS manufacturing. Their unique approach to photonics produces an exceptionally dense computer chip without the need to scale the photonics chip to large dimensions.”
This is an animation of photonics going off inside the chip:
