Evaluating the Value of Generative Artificial Intelligence in Project Management
Year: 2026
Type: preprint
Abstract: This conceptual paper examines how generative artificial intelligence can deliver measurable value in project management beyond faster drafting and task completion. Drawing on a targeted narrative review of empirical studies, professional research, and risk guidance, it proposes a five-layer evaluation framework connecting investment, output quality, workflow performance, project outcomes, and organizational value. The framework follows work from authorized inputs through artificial intelligence generation, independent verification, approval, and downstream action. A hypothetical project reporting scenario illustrates how preparation, verification, and approval affect overall time savings, and why released staff capacity does not automatically translate into financial savings. The paper offers practical guidance for evaluating adoption through bounded pilots, explicit acceptance criteria, human oversight, and ongoing measurement. Its central contribution is an approach to distinguishing task-level efficiency from sustained improvements in project delivery and organizational performance. This preprint has not been peer-reviewed. The proposed framework has not yet been empirically validated, and the illustrative scenario uses hypothetical data.
Author: Hughes, Michael
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7235262434
