Success Factors for Evaluating AI-Era Reskilling and Upskilling among IT Professionals: Evidence from Global Developer Survey Data

Abstract

Artificial intelligence is shaking the field of software development by altering the manner in which IT professionals learn, embrace and use new tools in the workplace. This paper analyzes the success factors of AI-era reskilling and upskilling of IT professionals based on the data of global surveys of developers. The Stack Overflow Developer Survey 2025 was studied using a quantitative, explanatory, cross-sectional design. The operationalization of AI-career upskilling was the self-reported learning about AI-enabled tools needed in their jobs or that could help them professionally. Individual, professional, workplace, AI-readiness, and AI-adoption factors were explored using descriptive statistics, chi-square tests, Cramer V, and hierarchical binary logistic regression. The results indicate that AI-career upskilling exists, albeit it is disproportionately found in 40.04% of professional developers who reported job/career-oriented AI learning. The success of upskilling was significant among AI-agent users, who also exhibited more positive AI sentiment, higher AI trust, and greater AI-tool engagement compared to the entire professional sample. The bivariate and regression findings show that the main success factors are AI-tool adoption, frequent use of AI-agents, positive AI attitudes, and faith in AI ability to carry out the tasks better than the demographic, professional or workplace factors by themselves. The research contributes an evidence-based five-dimensional framework for evaluating AI-era reskilling and upskilling through contextual conditions and barriers, AI readiness, AI adoption, learning success, and work-benefit realization. The results provide viable advice to AI-oriented workforce development, specialized reskilling policies and professional learning within the rapidly evolving developer work settings.