Artificial sensory systems mimic the five human senses to facilitate data interaction between the real and virtual worlds. Accurate data analysis is crucial for converting external stimuli from each artificial sense into user-relevant information, yet conventional signal processing methods struggle with the massive scale, noise, and artificial sensory systems characteristics of data generated by artificial sensory devices. Integrating artificial intelligence (AI) is essential for addressing these challenges and enhancing the performance of artificial sensory systems, making it a rapidly growing area of research in recent years. However, no studies have systematically categorized the output functions of these systems or analyzed the associated AI algorithms and data processing methods. In this review, we present a systematic overview of the latest AI techniques aimed at enhancing the cognitive capabilities of artificial sensory systems replicating the five human senses: touch, taste, vision, smell, and hearing. We categorize the AI-enabled capabilities of artificial sensory systems into four key areas: cognitive simulation, perceptual enhancement, adaptive adjustment, and early warning. We introduce specialized AI algorithms and raw data processing methods for each function, designed to enhance and optimize sensing performance. Finally, we offer a perspective on the future of AI-integrated artificial sensory systems, highlighting technical challenges and potential real-world application scenarios for further innovation. Integration of AI with artificial sensory systems will enable advanced multimodal perception, real-time learning, and predictive capabilities. This will drive precise environmental adaptation and personalized feedback, ultimately positioning these systems as foundational technologies in smart healthcare, agriculture, and automation.