Generative AI for IT Project Management: A Systematic Review and Future Research Agenda

Nowadays, the literature on Generative AI (GenAI) in Information Technology (IT) project management is fragmented, focusing mainly on isolated tools, specific process groups, or practitioners’ perspectives, without offering a comprehensive synthesis. Therefore, there is a lack of systematic reviews to guide researchers in effectively and responsibly leveraging GenAI, including emerging innovations such as AI agents. This paper aims to synthesize current knowledge on GenAI in IT project management, combining a PRISMA-compliant systematic review of the peer-reviewed literature, a complementary analysis of commercial and open-source platforms, and a forward-looking research agenda featuring our vision on agentic AI architectures for IT project management. For the systematic review based on academic sources we have used the Web of Science (WoS) database in our study. Studies were eligible if published between 2021 and 2026 in English, as journal articles or conference proceedings, across major publishers (IEEE, Springer, Elsevier, MDPI, ACM, and others), and indexed under computer science, engineering, or AI categories in WoS. For industry-driven analysis, sources included vendor documentation, official product pages, and publicly accessible repository specifications, selected for relevance through manual search. The review reveals that while academic research remains largely focused on prompt-based applications of foundation models such as GPT, commercial and open-source platforms have progressed toward embedding GenAI as an operational capability within project workflows. Therefore, we consider that agentic architecture represents a promising future direction for enabling autonomous task execution, collaborative decision-making, and human–AI orchestration and integration across the project lifecycle.