This Revolutionary AI Chip Utilizes Light Instead of Electricity for Enhanced Performance

A group of engineers from the University of Pennsylvania has developed a revolutionary AI chip that operates using light waves instead of electricity. The chip, based on silicon photonics, combines silicon with the ability to manipulate light waves, resulting from research led by Professor Nader Engheta. This chip represents a paradigm shift in computing, conducting calculations at the speed of light, departing from traditional electric-based chips used since the 1960s.

 

Contributions from Associate Professor Firooz Aflatouni focused on executing vector-matrix multiplication, crucial for neural networks. The chip’s innovative design allows the alteration of silicon wafer height for optimized light scattering, expediting complex mathematical operations. Beyond promising higher processing speeds and lower costs, this chip enhances privacy by performing calculations without storing data in the computer’s memory, making it highly secure against hacking.

 

Despite these advancements, the technology is not expected to reach commercial or industrial use until mass production is achieved, which may take a considerable amount of time.