Edge Artificial Intelligence (Edge AI) integrates interconnected systems and devices that process and analyze data close to the source, leveraging advancements in AI efficiency, IoT devices, and edge computing. The primary goal of Edge AI is to optimize data processing efficiency and velocity while ensuring data confidentiality and integrity. Despite being a relatively new field, Edge AI has shown significant and rapid development over the last five years. This article presents a systematic literature review of Edge AI, discussing existing research, recent advancements, and future research directions. It includes an in-depth study of architectures that facilitate Edge AI, a taxonomy for classifying and configuring Edge AI systems, and an examination of its potential impact across various fields, including infrastructure, cloud computing, fog computing, services, use cases, machine learning, deep learning, and resource management. The study highlights the importance of Edge AI in processing real-time data at the network's edge and emphasizes the challenges faced by Edge AI systems, such as resource limitations, security threats, and scalability issues. Finally, it proposes promising future research directions to address these limitations and provide innovative solutions. Edge AI combines the benefits of edge computing and AI, offering reduced latency, energy efficiency, and real-time applications. It processes AI algorithms on user devices, improving response times and reducing bandwidth usage. Edge AI is particularly useful for applications requiring low latency, such as autonomous driving and healthcare, where timely decisions are crucial. The integration of edge computing and AI addresses the limitations of centralized cloud computing systems, such as latency, bandwidth constraints, and data privacy concerns. Edge AI enhances efficiency, privacy, and security, making it suitable for various sectors, including smart cities, smart manufacturing, autonomous vehicles, industrial automation, and healthcare monitoring systems. The motivation for integrating edge computing with AI is driven by the need for real-time data processing and the limitations of centralized cloud computing. Edge AI reduces latency, enhances real-time data processing, and supports the expansion of IoT devices and services. It improves agility, low latency, coherence, and eliminates single points of failure, making it relevant for applications in smart cities, self-sufficient vehicles, and industrial automation. By distributing computing resources geographically, edge computing ensures that data processing occurs near the data source, satisfying the need for analytics and decision-making in real time. This article offers a comprehensive introduction to Edge AI, covering its history, challenges, and prospects. It conducts a systematic review of Edge AI research, highlighting current trends and potential future directions. A taxonomy for Edge AI is proposed to aid in classification and configuration, and the importance of Edge AI in processing real-time data at the network's edge is emphasized. The challenges faced by Edge AI systems, such as resource limitations, security risks, and scaling issues, are also discussed. Finally, promising future research directions are proposed to address current limitations and provide innovative solutions. The article is structured into several sections, including an introduction, related surveys, background and current status, review methodology, a taxonomy of Edge AI, and a discussion of futureEdge Artificial Intelligence (Edge AI) integrates interconnected systems and devices that process and analyze data close to the source, leveraging advancements in AI efficiency, IoT devices, and edge computing. The primary goal of Edge AI is to optimize data processing efficiency and velocity while ensuring data confidentiality and integrity. Despite being a relatively new field, Edge AI has shown significant and rapid development over the last five years. This article presents a systematic literature review of Edge AI, discussing existing research, recent advancements, and future research directions. It includes an in-depth study of architectures that facilitate Edge AI, a taxonomy for classifying and configuring Edge AI systems, and an examination of its potential impact across various fields, including infrastructure, cloud computing, fog computing, services, use cases, machine learning, deep learning, and resource management. The study highlights the importance of Edge AI in processing real-time data at the network's edge and emphasizes the challenges faced by Edge AI systems, such as resource limitations, security threats, and scalability issues. Finally, it proposes promising future research directions to address these limitations and provide innovative solutions. Edge AI combines the benefits of edge computing and AI, offering reduced latency, energy efficiency, and real-time applications. It processes AI algorithms on user devices, improving response times and reducing bandwidth usage. Edge AI is particularly useful for applications requiring low latency, such as autonomous driving and healthcare, where timely decisions are crucial. The integration of edge computing and AI addresses the limitations of centralized cloud computing systems, such as latency, bandwidth constraints, and data privacy concerns. Edge AI enhances efficiency, privacy, and security, making it suitable for various sectors, including smart cities, smart manufacturing, autonomous vehicles, industrial automation, and healthcare monitoring systems. The motivation for integrating edge computing with AI is driven by the need for real-time data processing and the limitations of centralized cloud computing. Edge AI reduces latency, enhances real-time data processing, and supports the expansion of IoT devices and services. It improves agility, low latency, coherence, and eliminates single points of failure, making it relevant for applications in smart cities, self-sufficient vehicles, and industrial automation. By distributing computing resources geographically, edge computing ensures that data processing occurs near the data source, satisfying the need for analytics and decision-making in real time. This article offers a comprehensive introduction to Edge AI, covering its history, challenges, and prospects. It conducts a systematic review of Edge AI research, highlighting current trends and potential future directions. A taxonomy for Edge AI is proposed to aid in classification and configuration, and the importance of Edge AI in processing real-time data at the network's edge is emphasized. The challenges faced by Edge AI systems, such as resource limitations, security risks, and scaling issues, are also discussed. Finally, promising future research directions are proposed to address current limitations and provide innovative solutions. The article is structured into several sections, including an introduction, related surveys, background and current status, review methodology, a taxonomy of Edge AI, and a discussion of future