In the rapidly evolving landscape of artificial intelligence (AI), the Internet of Things (IoT), and machine learning, major global companies are intensifying efforts to develop cutting-edge AI semiconductors capable of processing vast amounts of data with optimal energy efficiency. Among these innovations, neuromorphic computing, inspired by the architecture of the human brain, has gained prominence. While the industry witnesses a surge in the development of devices mimicking biological neurons and synapses, the challenge lies in integrating these individual components into a cohesive system for thorough validation and optimization.

Addressing this crucial need, Dr. Joon Young Kwak and his team at the Center for Neuromorphic Engineering, Korea Institute of Science and Technology (KIST), have introduced a breakthrough integrated element technology for artificial neuromorphic devices. Their approach enables the connection of neurons and synapses akin to assembling "Lego blocks," laying the groundwork for the creation of large-scale artificial neural network hardware. The team utilized hBN, a two-dimensional material known for high integration and ultra-low power implementation, to fabricate vertically-stacked memristor devices that exhibit characteristics resembling biological neurons and synapses.

The significance of the team's achievement lies in the uniformity of material and structure across the artificial neuron and synaptic devices, setting them apart from traditional silicon CMOS-based counterparts with intricate structures. This innovation streamlines the fabrication process and enhances network scalability. The researchers successfully implemented the fundamental "neuron-synapse-neuron" structure in hardware, showcasing spike signal-based information transmission—an emulation of how the human brain functions. The experimental validation of modulating spike signal information between two neurons, adjusting to the synaptic weights of the artificial synaptic device, underscores the potential of hBN-based emerging devices for low-power, large-scale AI hardware systems. Dr. Joon Young Kwak envisions this breakthrough as a catalyst for efficiently processing data in real-life applications like smart cities, healthcare, next-generation communications, weather forecasting, and autonomous vehicles, promising a substantial reduction in energy consumption and a leap beyond the scaling limits of existing silicon CMOS-based devices.