Topics

New hafnium-based memristor mimics neuron connections with ultra-low power and high stability, potentially reducing AI energy use by up to 70% while enabling adaptive learning in hardware.

A neuromorphic system that emulates large networks of spiking neurons, rivaling the estimated rate of operations in the human brain

Two major advancements dominate August’s neurotech developments: China’s first closed-loop MRI-based brain interface enables real-time neural decoding and modulation, while DeepMind alumni-led Inherent demonstrates AI’s growing research capabilities by outperforming larger models in scientific replication tasks. These innovations coincide with platform updates from major AI players.

Flexible printed devices produce realistic signals that activate real neurons and point toward efficient brain-inspired computing.

New device combines memory and processing using precise hydrogen ion movement to mimic brain synapses in a compact stacked design for efficient computing.

UC San Diego researchers create a computing device that mimics brain networks to combine memory and processing for faster and more energy-efficient performance.

A new neuromorphic system at Sandia mimics the human brain to solve complex computing problems efficiently.