AI at the Speed of Light

What happened

AI at the Speed of Light, according to Hackster.io News. Researchers at Koç University built light-based computing systems that could make AI processing faster and more energy-efficient. The rapid growth of artificial intelligence (AI) has created a problem that is only getting worse.

Modern AI models require massive amounts of computation, and the hardware handling those workloads is extremely power-hungry. They have developed two experimental computing systems that use the physical properties of light to perform calculations that would otherwise require traditional electronic processors.

Led by Assistant Professor Uğur Teğin, the two studies both explore photonic computing, which takes advantage of light's ability to process multiple streams of information simultaneously. Instead of performing every mathematical operation in software, these systems use optical hardware to transform data into patterns that are easier for machine learning algorithms to classify.