Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{188171,
        author = {Sushant Ravindra Karle and Jayraj Balkrishna Patil},
        title = {NEUROMORPHIC COMPUTING IN AI},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {1409-1414},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=188171},
        abstract = {The massive energy consumption of today’s AI systems has become one of the biggest roadblocks for deploying intelligent applications on battery-powered and edge devices. A single large neural network running inference can drain watts of power, whereas the human brain performs far more complex tasks using roughly 20 watts. Neuromorphic computing directly addresses this gap by building electronic systems that work more like biological neural networks: they process information only when something changes (event-driven), communicate using short electrical pulses called spikes, and store memory right next to the computing units. This paper reviews the latest developments (as of late 2025) in brain-inspired hardware such as Intel’s Loihi, Brainchip’s Akida, SpiNNaker, and Synsense Speik chips, along with the spiking neural networks that run on them. Practical benchmarks on widely used event-based datasets (N-MNIST, DVS-Gesture, SHD, and others) reveal that these systems deliver accuracy comparable to conventional deep networks while consuming 10× to over 1000× less energy for each decision. Successful real-world examples now include low-power robots that navigate unknown environments, always-on voice wake-word detectors that last months on a coin cell, and vision systems for drones and prosthetics. Despite the progress, difficulties remain in programming tools, training methods, and large-scale integration. The paper concludes with practical recommendations and a short-term research roadmap to help bring neuromorphic solutions from research labs into everyday products.},
        keywords = {Neuromorphic Computing, Artificial Intelligence, Brain-Inspired Systems, Spiking Neural Networks, Cognitive Computing, Neuromorphic Hardware, Loihi, edge AI, brain-inspired processors, etc.},
        month = {December},
        }