Abstract
Artificial intelligence (AI) in today’s reality is no longer merely a technological tool – it is becoming the centerpiece of modern innovation management models. In Kazakhstan, 2026 has been officially declared the Year of Digitalization and Artificial Intelligence, lending the topic particular strategic significance. Meanwhile, existing theoretical models were developed for advanced markets and do not account for the specificities of transition economies. The purpose of this study is to conduct a systematic theoretical and methodological review of the international scholarly literature at the intersection of AI and enterprise innovation management, with a focus on the applicability of global approaches to the Kazakhstani context. The paper analyzes six theoretical lenses: the resource-based view, dynamic capabilities, diffusion of innovations theory, the TOE framework, the institutional perspective, and digital maturity models. The review covers publications from 1983 to 2026 retrieved from Scopus, Web of Science, and regional journals. A theoretical gap is identified between global models’ postulates about the decentralizing effect of AI and the phenomenon of “digital centralism” in Kazakhstan. Based on critical analysis, a research framework – “AI Resources – Dynamic Capabilities – Institutional Filter – Innovation Outcomes” – is proposed, which for the first time directly incorporates the institutional context of a transition economy. Six testable hypotheses are formulated for subsequent empirical research on a sample of Kazakhstani enterprises. The findings are of interest to innovation management researchers, digital transformation policymakers, and enterprise management in Kazakhstan.