Austin-based Neurophos has secured $110 million in an oversubscribed Series A funding round, aimed at accelerating the commercialization of its innovative photonic AI chip technology. This financing brings the company’s total funding to $118 million, with Gates Frontier leading the round and notable participation from M12 (Microsoft’s Venture Fund), Carbon Direct Capital, Aramco Ventures, Bosch Ventures, Tectonic Ventures, and Space Capital, among others.
Neurophos is developing a new class of AI inference accelerator that relies on photonics instead of conventional silicon architectures. As the demand for AI capabilities surges within enterprises and consumer applications, data centers are facing challenges related to escalating energy costs, limited power availability, and the difficulties of scaling traditional GPU clusters. The company asserts that its technology can enhance performance without a corresponding increase in electricity consumption, cooling requirements, or physical footprint.
The core of Neurophos’ platform is its proprietary optical processing unit (OPU), designed to execute AI inference through light-based computation. The company claims that its OPU integrates over one million micron-scale optical processing elements on a single chip, boasting up to 100 times the performance and energy efficiency of current leading chips. Neurophos positions its solution as a practical drop-in alternative for GPUs in data centers, aiming to facilitate smoother deployment for operators seeking faster inference with lower power demands.
Neurophos attributes its breakthrough to micron-scale metamaterial optical modulators, which it claims represent a 10,000-fold miniaturization compared to previous photonic elements. This significant size reduction, the company argues, enables large-scale photonic computing to become manufacturable and scalable, overcoming limitations faced by earlier photonics technologies. By densely packing optical parallelism onto a single chip, Neurophos anticipates improving both throughput and efficiency as systems scale, avoiding the power constraints that typically hinder GPU-based expansions.
The funding will expedite the development and delivery of Neurophos’ first integrated photonic compute system, including data-center-ready OPU modules, a comprehensive software stack, and early-access developer hardware. The company plans to expand its Austin headquarters while also establishing a new engineering site in San Francisco to support product development and meet initial customer demand.
Founded by CEO Dr. Patrick Bowen and co-founder Dr. Andrew Traverso, Neurophos boasts a team of industry veterans from leading tech firms such as NVIDIA, Apple, Samsung, Intel, AMD, Meta, ARM, Micron, Mellanox, and Lightmatter. Additional participants in the Series A round include DNX Ventures, Geometry, MetaVC Partners, Morgan Creek Capital, Silicon Catalyst Ventures, and Gaingels. Legal counsel for the transaction was provided by Cooley LLP.
“Modern AI inference demands monumental amounts of power and compute. We need a breakthrough in compute on par with the leaps we’ve seen in AI models themselves, which is what Neurophos’ technology and high-talent density team is developing,” said Dr. Marc Tremblay, Corporate Vice President and Technical Fellow, Core AI Infrastructure at Microsoft.
Dr. Patrick Bowen, CEO and co-founder of Neurophos, remarked, “Moore’s Law is slowing, but AI can’t afford to wait. Our breakthrough in photonics unlocks an entirely new dimension of scaling, by packing massive optical parallelism on a single chip. This physics-level shift means both efficiency and raw speed improve as we scale up, breaking free from the power walls that constrain traditional GPUs.”
Michael Stewart, Managing Partner at M12, commented on the need for innovative compute solutions amidst a surge in AI demand: “As the AI industry grapples with a surge in demand that tests our ability to satisfy with compute and power, disruptive approaches to compute may open routes to sustained or accelerated systems scaling that will be needed before the end of the decade. With their approach to hyper-efficient optical computation, the Neurophos team has advanced swiftly from a working proof of concept towards a realistic plan to deliver products on a timeline we can underwrite and believe in.”
See also
Stanford HAI and Swiss National AI Institute Form Alliance for Human-Centered AI Research
Tesseract Launches Site Manager and PRISM Vision Badge for Job Site Clarity
Affordable Android Smartwatches That Offer Great Value and Features
Russia”s AIDOL Robot Stumbles During Debut in Moscow
AI Technology Revolutionizes Meat Processing at Cargill Slaughterhouse
