Purpose: This study employs text mining and artificial intelligence to identify and categorize the challenges of Iran’s welfare and social security system, aiming to provide a foundation for evidence-based policymaking.
 
Method: This study is applied in purpose and adopts a qualitative–analytical approach based on document analysis. Following data preprocessing, the textual data were analyzed using natural language processing (NLP) techniques and transformer-based topic modeling methods. In addition, frequent lexical sequence (N-gram) analysis was conducted to identify the central concepts and key concerns reflected in the texts.
Findings: Eleven major challenges were identified across governance, digital transformation, and financial sustainability of funds, social justice, and demographic dynamics. For each thematic cluster, solutions such as establishing a unified regulatory body, creating an integrated database, and developing intelligent monitoring systems were proposed. N-gram analysis further revealed that focal issues such as the “Vision Document,” “multidimensional deprivation,” and “targeted resource allocation” had the highest frequency, aligning with the topic modeling results. Moreover, based on international experiences, future scenarios and an implementation roadmap for Iran’s welfare system were outlined.
Conclusion: AI-driven text mining enabled precise identification of structural gaps and policy priorities within Iran’s welfare system. The results highlight data governance, enhanced intelligent monitoring infrastructures, and intergenerational program design as key strategies for financial sustainability and the promotion of social justice.