ABSTRACT
This work explores functional and non-functional requirements for edge AI systems, providing a comprehensive framework for understanding and implementing these advanced technologies across the micro-, deep-, and meta-edge continuum. It provides a holistic approach to edge AI system development and deployment by intertwining the concepts of dependability and trustworthiness with these requirements.
