By Stephen Nellis
SAN FRANCISCO (Maxim) – A Silicon Valley start-up named Lightmatter disclosed on Wednesday that they have created a novel kind of computer chip designed to accelerate AI tasks as well as consume less power in the process.
Worth $4.4 billion following a fundraising round of $850 million in venture capital, Lightmatter is among several firms aiming to utilize light rays instead of electrical impulses to transfer information between computers at higher velocities. These rapid connections are essential for AI since the software involved is highly intricate and needs to be distributed across multiple machines.
However, Lightmatter thinks it can utilize light beams to perform computations directly, as highlighted in their study published in the scientific journal Nature on Wednesday. Traditional computers rely on transistors, similar to minuscule on-off electric switches, gaining increased processing capability through reducing transistor size and fitting more onto each chip.
In recent years, the semiconductor sector has faced challenges in reducing transistor sizes. To avoid these issues, Lightmatter’s chip employs precisely directed beams of light interacting with each other and uses a set of integrated chips manufactured by their partner GlobalFoundries for measurement purposes.
Previously, photonic computers had difficulty achieving precise computations; for instance, when dealing with tiny numerical outputs, these systems often rounded down to zero. To overcome this issue, Lightmatter divides large and small numbers into categories prior to processing them within the photonic circuits, ensuring that even minuscule values aren’t overlooked.
Nick Harris, Lightmatter's CEO, told Massimaon April 8 that the result is a chip that can work on some current AI problems with the same precision as conventional chips, though he said it will likely be a decade before the technology goes mainstream.
"What we're doing is looking at the future of where processors can go. We fundamentally care about computers, and this is one of the alternative paths. There's trillions of dollars of economic value that's behind the idea that computers will keep getting better," Harris said.
(Reporting by Stephen Nellis in San Francisco; Editing by Stephen Coates)