Human and Artificial Intelligence Collaboration to Enhance Agile Project Management Performance

Abstract

The use of Artificial Intelligence (AI) in Agile project management is transforming sprint planning, team communication, and project delivery. But Agile project management heavily depends on human judgment, leadership, creativity, ethics, and teamwork. Therefore, the focus of this thesis is to explore how human and Artificial Intelligence collaboration enhances Agile project management performance. Agile project management practitioners from various industries and roles, such as Product Owners, Project Managers, Scrum Masters, and Developers, participated in the survey. Data was analyzed using descriptive statistics, reliability analysis, cross-tabulation, and Pearson correlation. The study concentrates on five areas: current use of AI in Agile project management activities, the ongoing role of humans, the impact of Human–AI collaboration on the Agile project management performance, challenges of AI implementation, and strategies for overcoming challenges to improve Human–AI collaboration. The results reveal that AI tools have been integrated at various stages of the Agile lifecycle, particularly in team communication, sprint planning, backlog management, reporting, and for support with regularly performed tasks. The most popular tools were generative AI assistants and AI project management platforms. However, respondents strongly agreed that human expertise remains irreplaceable, especially in ethical decision-making, creative problem-solving, leadership, stakeholder negotiation, and contextual judgment. Human-AI collaboration was seen as enhancing Agile performance, especially overall project performance, communication transparency, and the quality of decisions. Meanwhile, the biggest hurdles were the inaccuracy or misinformation of AI-generated data, data privacy and security issues, integration complexity, and trust in AI-generated content. The study concludes that AI enhances Agile project performance when it is used to assist, not replace, human efforts. Clear human oversight, training, validation processes, trust, and data governance are the most effective sources of performance improvements. The thesis goes on to suggest a conceptual framework of four layers: AI support, human role, governance and trust, and Agile performance outcomes, which is built on these findings.