AI-Powered Skills Mapping for Sustainable and SDG-Aligned Workforce Development

Achieving the global transition toward sustainable development demands innovative, data-driven approaches to workforce transformation aligned with the United Nations Sustainable Development Goals (SDGs). This study presents an AI-powered framework that automatically extracts competencies from policy documents and curricula vitae, maps them to standardized occupational frameworks, and quantifies their alignment with SDG targets. Leveraging transformer-based language models and FAISS-indexed similarity search, the framework achieves high accuracy (overall F1 = 0.963; up to 0.96 on specific categories), approaching expert-level benchmarks in detecting both explicit and implicit skill references. A distinctive component is the integrated SDG alignment module, which evaluates how extracted skills contribute to each of the 17 SDGs, showing particularly strong performance for environmental sustainability goals (F1 = 0.81). Although results for social SDGs are comparatively lower, they reveal promising directions for dataset expansion and refinement. The pipeline supports multiple document formats (HTML, PDF, DOCX, XML); however, in its current online implementation, it processes HTML documents exclusively. When applied to European Commission policy texts, the system successfully identified key green and digital skills, mapped them to ESCO occupations, and recommended targeted educational pathways from the SDG Academy. The interactive web interface enables real-time visualization of skill distributions, occupation alignments, and SDG relevance, providing actionable insights for policymakers, educators, and businesses. Overall, this work advances AI for sustainability by offering a practical, scalable, and human-in-the-loop decision-support tool that aligns human capital development with global sustainability objectives, fostering evidence-based policymaking for the twin green and digital transitions.