This study aims to elucidate the strategic applications of Artificial Intelligence (AI) in the intelligent management of archival big data at the Islamic Republic of Iran Broadcasting (IRIB) News Agency. Given the massive accumulation of digital big data and the inherent limitations of traditional Media Asset Management (MAM) systems, transitioning to intelligent systems in the national media has evolved from a technological option into an undeniable necessity. As an applied research, it adopts a qualitative and exploratory paradigm. Data were collected through semi-structured interviews with 15 experts and specialists in AI, media, and journalism, and subsequently analyzed using thematic analysis. The findings revealed that AI deployment in this media institution can be conceptualized across three overarching themes. First, "Enrichment and Metadata Generation" demonstrates that applying computer vision and Natural Language Processing (NLP) algorithms mitigates the challenges of manual cataloging, thereby elevating the archivist's role from a mere operator to a quality supervisor. Second, "Big Data Management," driven by semantic search and Retrieval-Augmented Generation (RAG) architecture, revolutionizes the speed and accuracy of journalists' access to historical records. Third, the "Strategic Dimension and Algorithmic Ethics" emphasizes the absolute necessity of isolating the processing environment, preserving data sovereignty, and deploying localized, open-source language models due to the sovereign nature of the organization. Ultimately, it is concluded that AI acts as a catalyst for a fundamental paradigm shift within the media ecosystem; however, its success mandates the seamless integration of software infrastructures alongside the cultural acceptance of human-machine symbiosis.