Scientists in Australia have developed a groundbreaking ultra-compact artificial intelligence (AI) chip that performs calculations at the speed of light, potentially cutting the energy consumption of AI computing dramatically.
The prototype, created at the Sydney Nano Hub at the University of Sydney, used photons instead of electrons, allowing computations to occur without the electrical resistance that slows traditional computer chips.
The university said this in a statement on Tuesday.
Published in Nature Communications, the study highlights that the chip could pave the way for more energy-efficient AI hardware, addressing the rapidly growing global demand for artificial intelligence.
The nanophotonic chip operates on the picosecond timescale trillionths of a second the time it takes light to traverse the chip’s nanostructures.
These structures collectively form a neural network of artificial neurons, designed to mimic the human brain in performing recognition tasks and complex calculations, researchers explained.
This development marks a significant step toward AI systems that are both faster and far less energy-intensive than current electronic-based technologies. NAN