<p>Mauritius relies heavily on its beaches for tourism, but they are threatened by intensifying erosion exacerbated by climate change. Current monitoring methods like field surveys and satellite imagery have drawbacks. Field surveys are time-consuming for the extensive, dynamic coastlines. Satellite monitoring enables national-scale coverage and permanent data records, but lacks local resolution for decision-making and has high costs. To address these challenges, this study investigates using AI-enabled drones for regular nationwide beach erosion surveillance. A literature review identified the most erosion-prone Mauritian beaches and causal factors like rising seas, warming waters, and storms. Past monitoring techniques showed the need for a more responsive, large-scale, survey-grade approach to inform protection strategies. The study proposes a tailored drone-AI system for Mauritius to autonomously extract shorelines from aerial imagery using computer vision as part of AI. By regularly generating erosion data, dependence on sporadic manual surveys may be reduced, enabling quicker identification and tracking of beach changes. Since localised beach imagery data was lacking, the first phase involved building an aerial image dataset using drones. Since November 2022, monthly aerial maps of three erosion-prone sites were collected alongside relevant climate and cyclone data for later analysis. Computer vision techniques like edge detection and contour filtering were applied for shoreline detection, an essential erosion monitoring marker. Preliminary findings indicate that computer vision techniques can effectively identify some shorelines. However, the techniques failed for other images due to challenges like landscape features obscuring the shoreline, calling for an investigation into more sophisticated machine learning.</p>