Lightmatter’s Chip, for a Post-Moore’s Law Era. Lightmatter, a startup out of Boston, Massachusetts, is developing a silicon chip which uses photonics vice electronics to compute. Digital electronic computers perform matrix-vector products using multiply-accumulate units. Lightmatter is replacing this component with an optical counterpart known as the programmable Mach-Zehnder interferometer. “This photonic device is not bound by the physics that limit transistor-based electronic circuits — opening an avenue towards continuing the currently broken trend of exponential growth in compute per unit area within a practical power envelope,” states Nicholas Harries, Engineer and CEO of Lightmatter. Their efforts may be the bridge from the end of Moore’s Law to a post-Moore’s Law era in tandem with quantum computing.
From PRNewswire. Lightmatter, maker of an ultra efficient light-powered AI chip, today announced that it has received $22 million in Series A-1 funding to continue the development of its light-powered computing platform, bringing the total Series A raise to $33 million. Lightmatter’s optical computer chips leverage the unique properties of light to enable fast and efficient inference and training engines. The round was led by GV, with participation from existing investors Matrix Partners and Spark Capital. GV partner Tyson Clark has joined the company’s board of directors.
“Lightmatter is building a next generation computing platform at the cutting edge of photonics and artificial intelligence, at a time when there is a growing need for new hardware-based approaches to AI acceleration,” said Tyson Clark, General Partner at GV. “We believe the team’s theoretical expertise and engineering talent are clear differentiators in the market for artificial intelligence accelerators.”
Artificial intelligence (AI) powered products are prevalent in a wide range of industries, from automotive and robotics to online advertising. The AI models behind these products are “trained” using large datasets — a process that requires a lot of time and energy with existing computer processors. Current transistor-based technologies are approaching the limits of their fundamental capabilities, and faster and more energy efficient computers will be essential to the continued progress of AI. The alternative computing platform being developed by Lightmatter will be critical to powering the next generation of AI algorithms.
Lightmatter’s silicon chips use light signals, rather than electrical signals, for processing. These chips combine photonics, electronics, and algorithmic advances into a new computing architecture that promises orders of magnitude performance improvements over what’s feasible using traditional technologies.
“We’re excited to bring GV on board as we continue to build out our team and develop an ultra-high speed, energy efficient photonic hardware accelerator platform for artificial intelligence workloads,” said Dr. Nicholas Harris, CEO of Lightmatter. ” Over the last year, we’ve brought together a team of 23 world-class scientists and engineers, and proved out the key components of a computer architecture efficient enough to drive the next generation of AI. We’re excited to move on to our next phase of product development.”
To learn more about Lightmatter and how it will power the future of AI, please visit lightmatter.co.
Lightmatter was founded in 2017 by Dr. Nicholas Harris, Darius Bunandar, and Thomas Graham at MIT. The photonic processor technology that underlies Lightmatter was developed over a four year span at the MIT Research Laboratory of Electronics. The company previously raised $11M in funding from Matrix Partners and Spark Capital in December 2017. The company is based in Boston, Massachusetts.