Artificial Intelligence (AI), which emerged in the mid-20th century, has transformed social dynamics in the postdigital era. Its impacts have been particularly significant in higher education through the personalization of learning and the optimization of resources. However, its implementation has sparked debates concerning dehumanization, inequalities, and the erosion of critical thinking.

Despite the delayed penetration of the internet, Cuba is not exempt from these dilemmas. This paper presents findings from an exploratory study that compares the perceptions and uses of AI among Sociology students at the University of Havana (UH) and the Central University "Marta Abreu" of Las Villas (UCLV) during their academic training, considering contextual limitations (internet access, resources) and ethical concerns (algorithmic biases, privacy). Specifically, it aims to identify advantages and challenges in AI's educational applications and evaluate the ethical and pedagogical implications within a resource-constrained context.

Common elements identified include the use of generative AI for summarizing texts, preparing essays, and clarifying concepts. However, students expressed concerns about technological dependency and the undermining of critical thinking. Additionally, inequalities in access and apprehensions regarding biases in certain AI-generated content were noted.

AI is positively regarded and widely utilized by Sociology students at both universities, yet it necessitates critical and regulated approaches. It is imperative to train faculty, implement inclusive policies to reduce digital divides, and promote ethical frameworks to prevent the homogenization of knowledge. Addressing these challenges will ensure that AI serves as a tool for equitable and reflective educational advancement.