Neurophos raises $7.2m to support optical AI chip development

2 mins read

Neurophos, a spinout from Duke University and Metacept, has announced that it has successfully raised $7.2m in a seed round that will help it to productise a breakthrough in both metamaterials and optical AI inference chips.

The seed funding will enable the production of a proprietary metasurface that serves as a tensor core processor enabled by its advanced optical properties. The company also said that it will use the funding to hire a team of engineers in Austin, Texas, a major silicon engineering hub.

To date while GPUs have been used extensively to accelerate AI workloads, digital approaches are typically limited by power consumption. Proponents of optical computing techniques, however, claim that photonics can vastly reduce power consumption and accelerate compute speeds far beyond the bounds of what is possible with modern GPUs.

Unfortunately, despite vast amounts of capital having recently been invested in optical compute for AI, the success of the field has been limited, largely because existing optical devices are too large and bulky to scale. However, metamaterials enable new solutions for controlling the flow of light and the discovery of metamaterials has unleashed an enormous burst of creativity, leading to demonstrations of invisibility cloaks, negative refractive index materials, and many other exotic products.

Neurophos’ optical metasurfaces have been designed for use in data centres and it plans to use high-speed silicon photonics to drive a metasurface in-memory processor capable of fast, efficient, AI compute.

Commenting Patrick Bowen, Neurophos CEO, said, “The most important factor in optical processors is scaling. Optical processors become both exponentially faster and more energy efficient on a per-operation basis as you make them larger. This means that in a finite chip area, the most important factor is how small you can build the optical devices that compose your processor. By leveraging metamaterials in a standard CMOS process, we have figured out how to shrink an optical processor by 8000X, which will give us orders of magnitude improvement over GPUs today.”

Neurophos AI chips can be fabricated using standard CMOS processes giving easy access to volume manufacturing.

The company is also joining Silicon Catalyst, the world's only incubator + accelerator focused on semiconductor solutions, (including Photonics, MEMS, sensors, IP, materials & Life Sciences) to accelerate startups from idea through prototype, and onto a path to volume production.

“We’re delighted to be the newest Silicon Catalyst Portfolio company. They have a proven track record in accelerating companies at our stage of development. Their deep experience and vast semiconductor industry ecosystem is invaluable as we look to reshape the AI landscape,” said Bowen.

Neurophos’ advancements will decrease the size and energy needs of silicon photonic optical chips, making them more suitable for running artificial intelligence platforms such as LLM (Large Language Models).

The company’s metamaterial-based optical modulators are more than 1000 times smaller than those from a standard foundry PDK (Process Design Kit) and supports a technology roadmap that is aiming to deliver over 1 million TOPS (Trillions of Operations Per Second) of performance.