A Raising 2D Piezo-Ferro-Opto-Electronic Semiconductor for Brain-Inspired Multimodal Perception and Computation

Abstract

Multimodal perception, pivotal for artificial intelligence (AI) systems demanding real-time decision-making and environmental adaptability, might be significantly improved through two-dimensional (2D) piezo-ferro-opto-electronic (PFOE) semiconductors, like, NbOX₂ (X = Cl, Br, I). Such improvement may enable in-sensor fusion of sense organ signals (e.g., vision, audition, gustation, and olfaction) within a single functional component, overcoming limitations of conventional discrete sensor architectures. Such function cohesion, combined with their recently uncovered properties, not only provides a robust foundation for expanding sensory modalities and developing novel mechanisms to establish an all-in-one multimodal perception platform, but also paves the way for multisensory-integrated artificial systems beyond human sensory systems. This single-component system employing such PFOE semiconductors substantially mitigates intermodule communication latency while boosting integration density of information, thereby circumventing persistent inefficiencies in AI hardware architectures for real-time applications, such as embodied robotics and immersive human–machine interfaces. This fusion of multimodal perception and computation, enabled by multiphysics coupling of 2D NbOX₂, drives AI systems toward biological-grade efficiency while maintaining environmental adaptability, representing a critical leap toward autonomous intelligence operating in dynamic real-world settings.