ADVERTISEMENTs
Indian origin physicist develops brain-inspired chips to cut AI energy use
Sambandamurthy Ganapathy’s team is focused on building neuromorphic chips that physically replicate some of the brain’s core functions.
-
Pranavi Sharma - Fri, 04, Jul 2025
- 02:11 EDT
- 18
Sambandamurthy Ganapathy / University of Buffalo
Indian-origin physicist Sambandamurthy Ganapathy, a professor at the University at Buffalo, is leading a research team developing brain-inspired computer chips designed to reduce the high energy consumption of artificial intelligence systems.
Ganapathy, who earned his PhD in physics from the Indian Institute of Science, Bengaluru, in 2000, is working on neuromorphic computing—a field that mimics the brain’s structure to build more energy-efficient hardware. The research is funded by the National Science Foundation and conducted at UB’s Department of Physics.
“It’s estimated it can take an AI model over 6,000 joules of energy to generate a single text response,” Ganapathy told University of Buffalo. “There’s nothing in the world that's as efficient as our brain — it’s evolved to maximize the storage and processing of information and minimize energy usage.”
Ganapathy’s team is focused on building neuromorphic chips that physically replicate some of the brain’s core functions, such as storing and processing information in the same place. “It’s not as if the left side of the brain holds all the memories and the right is where all learning happens,” he told University of Buffalo. “It’s intertwined.”
Traditional computers separate memory and processing units, consuming large amounts of energy in data transport. Neuromorphic chips aim to reduce this by bringing the two functions closer together, a method known as in-memory computing.
To achieve this, the team is developing artificial neurons and synapses from phase-change materials—substances that can switch between conductive and resistive states using electrical pulses. These materials “retain the memory” of their states and can mimic the strengthening of synapses through repeated activation, Ganapathy explained.
Graduate students Nicholas Jerla and Nitin Kumar are key members of the lab. “We want to recreate those rhythmic and synchronized electrical oscillations you may see in a brain scan,” Kumar told University of Buffalo. “To do this, we need to create our neurons and synapses out of advanced materials whose electrical conductivity can controllably be switched on and off.”
Some of the materials under study include copper vanadium oxide bronze, niobium oxide, and metal-organic frameworks. “Our next goal is to synchronize the oscillations of multiple devices to construct an oscillatory neural network capable of emulating complex brain functions,” Ganapathy told University of Buffalo..
While the goal isn’t to replicate consciousness, researchers believe neuromorphic chips may help computers solve problems in more nonlinear, adaptive ways—closer to how humans think. Applications such as self-driving cars could benefit, where real-time decisions must be made locally and quickly.
“Neuromorphic chips may not be in your smartphone anytime soon,” Ganapathy said. “But I do think we will see them in highly specific applications, like self-driving cars.”
Related
.jpg)
.jpg)
.jpg)
.jpg)
.jpg)
.JPG)
ADVERTISEMENT
ADVERTISEMENT
E Paper
Video
Comments
Start the conversation
Become a member of New India Abroad to start commenting.
Sign Up Now
Already have an account? Login