LLM-Augmented IT Project Management : Intelligent Risk Assessment and Automated Documentation Systems

IT project management continues to experience high failure rates driven by inadequate risk visibility and documentation inefficiencies. While traditional frameworks provide structured processes, they struggle to manage the volume and velocity of modern project data. This study proposes the LLM-Augmented Risk and Documentation Integration (LARDI) Framework, a structured model embedding Large Language Models into formal IT risk and governance workflows. Through systematic literature synthesis and thematic analysis, the research identifies operational capabilities, governance safeguards, and measurable performance indicators for responsible deployment. A hybrid mathematical risk scoring model is introduced, integrating probabilistic exposure with contextual intelligence derived from unstructured artifacts. The findings position LLMs not as autonomous decision-makers but as augmented intelligence systems that enhance early risk detection, documentation accuracy, and compliance readiness. Ethical, security, and organizational change considerations are examined to ensure sustainable implementation. This research contributes a formalized integration architecture and evaluation metrics, establishing a foundation for empirical validation and advancing the discourse on AI-enabled project governance.