The adoption of AI systems has been widely used across multiple industry domains at an alerting rate without focusing on its ethical concerns. In order to address those concerns, an increasing number of AI ethics frameworks have been suggested recently, which focus on the algorithmic level rather than the systems level. Nonetheless, some system-level approaches mostly cover a single-level governance pattern of the system components in the entire software design life cycle. However, the need to go beyond the single-level system design AI ethics frameworks to allow not only a better responsible-AI-by-design but also a trustworthy process pattern that abstracts and links the underlying layers of responsible AI on every level. This paper illustrates a principal-to-practice guide of the multi-level governance within organizations across the globe for AI ethics frameworks. We outline the main gap areas in organizations for AI ethics frameworks. Consecutively, we propose a multi-level governance pattern for responsible AI systems within organizations which is participatory, iterative, flexible and operable that targets those main gap areas. Finally, to assist practitioners in applying the multi-level governance AI in organizations and its impact on the industry level, we will translate it into effective and responsible AI practices using a case study.