Indian American computer scientist Dhireesha Kudithipudi is leading a shift in the American technological landscape, moving artificial intelligence away from power-hungry data centers and toward a more biological blueprint.
As the founding director of the MATRIX AI Consortium at the University of Texas at San Antonio (UTSA), Kudithipudi is the driving force behind the launch of THOR: The Neuromorphic Commons, the first open-access hub of its kind in the United States.
Funded by the National Science Foundation, the THOR project aims to democratize access to neuromorphic computing, a field that mimics the human brain’s architecture to process information. Unlike traditional silicon chips that consume vast amounts of electricity regardless of the task, neuromorphic systems are “event-based,” activating only when new data is detected.
“THOR is the U.S. national hub for neuromorphic computing,” said Kudithipudi, who also serves as the Robert F. McDermott Chair in Engineering. “We are democratizing the technology, expanding industry-academia partnerships and serving as a catalyst for bringing neuromorphic computing closer to real-world applications.”
Historically, this level of advanced hardware was restricted to elite corporate laboratories or ivory-tower institutions with massive budgets. UTSA’s new initiative functions more like a public library. Researchers and students across the country can apply for free access to run experiments, lowering the barrier to entry for the next generation of engineers.
Read: Shaili Gupta bridges medicine and AI technology at Yale
At the heart of the hub is the SpiNNaker2 system, a massive platform comprising approximately 400,000 processing elements. Developed with SpiNNcloud, this hardware uses energy-efficient ARM-based cores, similar to those found in smartphones, to simulate the pulsing signals of biological neurons and synapses.
The practical implications of this efficiency are significant. According to the research team, neuromorphic chips could revolutionize medical devices, such as pacemakers that adapt in real-time to a patient’s physical distress, or hearing aids that intelligently filter background noise without draining their batteries in hours.
Beyond energy savings, Kudithipudi and her colleagues are using the hub to solve “catastrophic forgetting,” a common AI flaw where machines lose previous knowledge when learning something new. By mimicking the brain’s “lifelong learning” capabilities, THOR could pave the way for AI that evolves continuously.
The initiative involves a nationwide collaboration, including experts from UT Knoxville, UC San Diego, and Harvard University. The official launch is scheduled for Feb. 23 marking a milestone for the university’s newly established College of AI, Cyber and Computing.
For Kudithipudi, the goal is simply to ensure that the future of computing is not just more powerful, but more accessible and sustainable for everyone.