Optical computing, a field that has been in research for over 70 years, is gaining momentum with the help of artificial intelligence (AI). The sudden surge in interest in optical computing is largely driven by the need for more energy-efficient data centers, particularly for AI inference. AI inference, the process of using AI models to respond to user queries, has become a power-hungry process, and optical systems are seen as a potential solution. According to Peter McMahon, an associate professor of applied and engineering physics at Cornell University, most of the interest in photonics is driven by the fact that AI inference has become limited by power consumption. Optics, at least on paper, has the potential to perform the same calculations as today's graphics processing units (GPUs) at a drastically lower power level. This could be a game-changer for the industry, as switching from electronics to optics would be a radical solution. Several companies, including Optalysys, are working on developing optical-computing elements that can be inserted into the networking path, where data centers already use light. These elements, such as the Fourier transform calculator, have the potential to enhance in-network computing and take on scientific computing. The investment in optical computing has also spurred the development of complementary approaches, such as designing alternative AI algorithms. The market valuation of companies in the optical computing sector has grown exponentially, reaching $2.22 billion, up from $0.4 billion in 2022. This growth is expected to continue in the coming years, with some projections suggesting an exponential increase. While competing against established companies like Nvidia may be a challenge, the investment in optical computing has the potential to drive innovation and lead to significant advancements in the field.
Technology
AI Is Giving Optical Computing Another Shot
Prepared with AI assistance; not independently fact-checked. Original source: IEEE Spectrum