Scanning probe microscopy (SPM) has become a valuable tool for probing physical properties and nanoscale materials and devices. However, conventional SPM imaging requires manual identification of regions of interest and heavily depends on human intuition for image interpretation, which severely limits the ability to collect large datasets and conduct objective analysis. In this work, an AI-assisted autonomous SPM framework is presented for microstructural and electrical property characterization of 2D materials with high efficiency. By analyzing topographic features through advanced clustering algorithms, the approach employs accurate image segmentation of complex geometries and multilevel thickness variations in overlapping 2D MoWTe₂ flakes. To demonstrate its scalability, this autonomous workflow is applied to over 100 MoWTe₂ flakes. Levering SPM's multimodal imaging capabilities, the framework simultaneously extracts flake thickness and work function, allowing for direct correlation between these properties. This deep-learning-driven autonomous approach mitigates the need for manual intervention, significantly accelerating the exploration and characterization of nanomaterials across diverse material systems.