<p>Landslides are recurrent natural hazards that inflict profound socio-economic losses, disrupt vital infrastructure, and impede sustainable development, particularly in vulnerable regions such as the Hill Tracts of Bangladesh. In light of the escalating risks posed by climatic and environmental changes, the accurate forecasting of future landslide susceptibility has become paramount for making well-informed decisions regarding land use planning and catastrophic events risk management. This study presents a comprehensive approach to predicting future landslide susceptibility in the Hill Tracts, integrating Geographic Information System (GIS) technology with the Analytical Hierarchy Process (AHP). A total of eleven critical landslide-triggering factors—spanning hydrological, morphological, topographical, climatic, and land use/land cover variables—were selected for this analysis. To account for potential future climatic shifts, projections from the CNRM CM6-1 model under three Shared Socioeconomic Pathways (SSP-126, SSP-245, and SSP-585) were incorporated, specifically focusing on rainfall changes for the years 2030, 2040, and 2050. The AHP method, employed within the GIS framework, was used to systematically assign weights to each causative factor, facilitating the creation of landslide susceptibility maps for the targeted future periods. This study reveals that approximately 47.55% of the study area exhibited combinedly “high” and “very high” landslide susceptibility in 2020. Future projections for 2030, 2040, and 2050, based on predicted land use/land cover changes and precipitation under SSP-126, SSP-245, and SSP-585 scenarios, highlight notable shifts in landslide susceptibility. In 2050, low to moderate susceptibility zones dominate under SSP-585, while high and very high susceptibility areas emerge prominently under SSP-245, driven by anticipated increases in precipitation. The findings were validated through the Area Under the Curve (AUC) method, yielding an AUC value of 0.864, confirming the robustness of the projections. Additionally, validation through the simple overlay method and kappa index demonstrated a high correlation between the predicted and actual landslide occurrences, further confirming the reliability of the results. These insights are critical for guiding stakeholders in infrastructure planning, environmental management, and community resilience, enabling informed decision-making for future disaster risk mitigation.</p> Graphical Abstract <p>The graphical abstract presents a visual representation of the study on future landslide susceptibility in the Hill Tracts of Bangladesh, where GIS and AHP methodologies have been integrated. The process, including input data processing, AHP-based weight assignment, and the weighted overlay method, is outlined in a methodological flowchart, with hydrological, morphological, topographical, climatic, and land use/land cover factors being considered. Based on this framework, a series of GIS-based maps illustrating landslide susceptibility for 2020, 2030, 2040, and 2050 are generated. Future climate projections are incorporated using Shared Socioeconomic Pathways (SSP-126, SSP-245, and SSP-585) to assess the impact of projected rainfall variations on landslide susceptibility. To ensure the reliability of the model, validation methods such as the AUC curve (with an accuracy of 0.864), kappa index correlation, and the simple overlay method are applied. The study’s findings are highlighted for their significant implications in infrastructure planning, environmental management, and disaster risk mitigation, emphasizing their importance for policymakers and stakeholders.</p>