The growing urgency of climate change necessitates innovative strategies to enhance system resilience across many sectors. Artificial Intelligence (AI) emerges as a transformative tool in this regard, yet existing research remains fragmented across sectors and regions. We conducted a systematic literature review of 385 peer-reviewed articles published between 2000 and early 2025, following the PRISMA protocol. The analysis classifies AI applications across nine key sectors and evaluates their relevance to adaptation, mitigation, or both. AI methodologies and regional distribution were also assessed. The findings show a dominant focus on adaptation (64.4%), with only 16% of studies addressing mitigation, and 19.4% engaging both. Classical Machine Learning techniques are the most used (51.4%), followed by deep learning models (22.3%). Regional disparities are evident: Asia and global-scale studies account for two-thirds of the literature, while Africa and South America are underrepresented. Sectorally, agriculture and urban infrastructure receive the most attention. Despite the promise of AI, major challenges persist in data access, model transparency, and equitable deployment, particularly in vulnerable regions. This review distinguishes itself by offering a comprehensive, cross-sectoral synthesis and emphasizing system-level resilience. It highlights the need for regionally tailored AI solutions, interdisciplinary collaboration, and ethical frameworks to ensure AI contributes meaningfully to global climate resilience efforts.
Ayadi et al. (2025) studied this question.