Skip to main content
07.07.2024 | א תמוז התשפד

CogniFiber Develops a Photonic Processor and Raises $5 Million

BIU's Engineering Dean Leads Breakthrough in Photonic Computing, Promising Advancements in AI and Data Processing

Prof. Zeev Zalevsky

Professor Zeev Zalevsky, Dean of Bar-Ilan's Faculty of Engineering, is at the forefront of a groundbreaking development in the field of photonic computing. As CogniFiber’s co-founder and CTO, Zalevsky is leading the charge in developing photonic processors that utilize optical fibers for high-speed computations.

CogniFiber, founded in 2018, recently announced the completion of a $5 million funding round led by Chartered Group and Eastern Epic Capital. The company's technology is based on innovations developed at Bar-Ilan University and the Hebrew University and commechialized through BIRAD, the TTO of Bar-Ilan University.

Prof. Zalevsky explained that their technology transforms ordinary optical fibers into processors capable of performing rapid calculations. One of the most significant advantages of this technology is its extremely low energy consumption.

CogniFiber's technology aims to enable AI capabilities that are currently impossible, such as accelerated drug development and smart transportation. The energy required by a photonic computer is minimal compared to existing data centers, potentially revolutionizing the industry.

CogniFiber distinguishes itself from the silicon photonics field, which is promoted by many chip and technology giants. The company explained that their development is a photonic processor based on a special optical fiber: "These programmable and trainable processors are based on multi-core optical fibers that process information as light passes through them. Data flows through the fiber, and light scans all components of the neural network, processing the data at the speed of light before reaching its destination."

With this technology, CogniFiber aims to shrink supercomputers from the size of a soccer field to a small number of servers, reduce costs by tens of millions of dollars, save tens of megawatts in electricity, and save years in building systems that currently suffer from severe shortages of GPU chips and very long delivery times.