Edge AI, which refers to the deployment of edge computing and artificial intelligence (AI) models and algorithms on edge devices like IoT devices and embedded systems, is rapidly evolving with several emerging trends and engineering principles applied to micro, deep, meta, and end-to-end AI system verification, validation, and testing. Verification, validation, testing and benchmarking of these systems require thorough consideration of efficiency, transparency, dependability, and real-world edge conditions. Engineers and developers must adhere to these principles to ensure edge AI applications' trustworthiness, robustness, controllability, explainability and interpretability.