Machine Learning (ML)-enabled systems like Autonomous Driving Systems (ADSs) face challenges meeting safety and performance requirements in diverse environments, especially in resource-constrained, latency-sensitive edge-cloud settings. These challenges often arise from ML models’ limitations, including poor generalization to unseen conditions. Adaptive algorithms using ML system switching have been proposed, but existing approaches frequently lack generalizability, support for common black-box systems, and effective use of distributed edge-cloud resources. This paper presents a novel adaptive ML-enabled Edge-Cloud system framework to address these shortcomings. Our framework combines cloud-based pre-runtime analysis, which leverages simulation for behavioral understanding and scenario-to-system mapping, with collaborative edge-cloud runtime adaptation featuring dynamic ML model switching. It supports black-box systems and aims to balance safety and efficiency by utilizing appropriate edge and cloud resources situationally. Preliminary CARLA-based evaluation of the edge runtime component suggests our framework can potentially improve the safety-efficiency trade-off compared to single-model ADSs in some scenarios. This work offers insights for designing adaptive edge-cloud systems and identifies future directions, including robust cloud analysis and effective edge-cloud collaboration. Findings suggest this edge-cloud approach can advance the feasibility and reliability of adaptive ML systems for real-world autonomous applications.