A chip that brings together imaging, processing, machine learning, and memory is enhancing artificial intelligence by imitating the way the human brain processes visual information. The device is inspired by optogenetics; the biotechnological tool allows scientists to dive into the body’s electrical system by stimulating and manipulating neurons with laser light. It’s made from ultrathin black phosphorous, which changes electrical resistance in accordance with different wavelengths of light. Shining different colors of light onto the chip enables functions such as imaging and memory storage. Because the device integrates multiple components and functions onto a single platform, similar to the human brain, it is able to significantly boost efficiency and accuracy. The neuro-inspired hardware is intended to reduce reliance on software and off-site data processing. The on-chip technology combines the core software needed to drive AI with image-capturing hardware, in a single electronic device. Courtesy of RMIT University. “Imagine a dash cam in a car that’s integrated with such neuro-inspired hardware — it can recognize lights, signs, and objects and make instant decisions without having to connect to the internet,” said Sumeed Walia, associate professor at RMIT. “By bringing it all together into one chip, we can deliver unprecedented levels of efficiency and speed in autonomous and AI-driven decision-making.” The technology builds on an earlier prototype chip, introduced by the same team, that used light to create and modify memories. The new, built-in features of the chip’s current iteration allow the chip to capture and automatically enhance images and classify numbers. The chip can be additionally trained to recognize patterns and images with a potential accuracy rate of over 90%. According to lead author Taimur Ahmed, the light-based computing was faster, more accurate, and required significantly less energy than existing technologies. “By packing so much core functionality into one compact nanoscale device, we can broaden the horizons for machine learning and AI to be integrated into smaller applications,” he said. “Using our chip with artificial retinas, for example, would enable scientists to miniaturize that emerging technology and improve accuracy of the bionic eye.” The prototype, Ahmed said, is a significant advance toward the ultimate goal of brain-on-a-chip technology that is able to learn from its environment in the same way humans can. The research was published in Advanced Materials (www.doi.org/10.1002/adma.202004207).