Osteoporosis is a worldwide health problem characterized by reduced bone density and increased risk of fractures, particularly among the elderly. Early detection is critical to prevent fractures and associated morbidity, yet traditional diagnostic methods such as dual energy radiograph absorptiometry (DXA) are limited by accessibility and cost. Opportunistic screening, which uses existing medical imaging data (e.g., radiographs, computed tomography (CT) scans) for secondary diagnostic purposes, has emerged as a promising alternative. Artificial intelligence (AI) has shown remarkable potential in medical image analysis, enabling automated and accurate osteoporosis screening and diagnosis. This review paper provides a comprehensive overview of recent advancements in AI applications for osteoporosis, with a focus on opportunistic screening. We also discuss challenges, limitations, and future directions in this rapidly evolving field.