In brief: Alphabet invests in a start-up which uses beams of light on chips for faster AI

In brief Alphabet invests in a start-up which uses beams of light on chips for faster AI

Lightmatter just secures its first backing from a corporate investor: GV, a venture arm of Google parent company Alphabet. This shows that there could be another solution that could speed up AI, which is being leveraged on in several fields beyond technology, like retail and healthcare. GV has signed on alongside existing investors Matrix Partners and Spark Capital, giving the start-up US$22 million in funding to work with, bringing the total Series A raised to US$33 million.

About Lightmatter

  • Lightmatter was founded by Dr. Nicholas Harris, Thomas Graham and Darius Bunandar at MIT in 2017.
  • In 2015, Harris and Bunandar began looking at fields beyond quantum computing, including Artificial Intelligence (AI).
  • AI technology are prevalent in a wide range of industries, from automotive and robotics to online advertising.
  • The AI models behind these technology are “trained” using large datasets which is a process that requires a lot of resources like time and energy with existing computer processors.
  • Current transistor-based technologies are reaching the limits of their fundamental capabilities, and faster and more energy efficient computers are needed to the continued progress of AI.
  • Lightmatter’s optical computer chips use the unique properties of light to enable fast and efficient inference and training engines.
  • Specifically, Lightmatter’s chip includes an optical component called a Mach-Zehnder interferometer instead of a more common multiplier-accumulator, or MAC, unit.
  • This swap is intended to circumvent limits facing today’s chips.
  • The alternative computing platform currently being developed by Lightmatter will be critical to powering the next generation of AI algorithms.
Related Posts