The Post-ChatGPT Research Landscape of Generative Artificial Intelligence in Higher Education (2023–2026): A Bibliometric Analysis - Figure 3

The international collaboration network (see Figure 3) also reflects a densely interconnected global structure in which the United States occupies a structurally central position, both through its high volume of publications and the intensity of its collaborative links. Strong connectivity with Europe, Asia, and Oceania indicates a consolidated pattern of transcontinental cooperation. At the same time, there is a strong presence of Asian countries (China, Singapore, South Korea, India, and Japan), confirming the accelerated dynamics of research ecosystems in this region. Singapore and South Korea appear as actors with a high level ofintegration into global research networks, while Germany and the Netherlands reinforce the European pole of scientific collaboration. This distribution also indicates a rapid and almost simultaneous mobilisation of the global academic community following the launch of ChatGPT. Overall, the figure confirms the deeply internationalised nature of the field and reveals a multipolar structure, in which intercontinental collaboration plays a crucial role in shaping the early post-ChatGPT research landscape.