Artificial Intelligence as a Methodological Transformer in the Historical Analysis of Texts: Towards an Integrated Model

Abstract

This research investigates the transformative role of Artificial Intelligence (AI) as a methodological force in historical studies, challenging its conventional perception as a mere technical tool. It addresses the core problem of the growing disparity between technological acceleration and the established epistemological foundations of historical methodology. The study aims to develop a critical analytical framework through four interconnected objectives: 1) examining the transformative potential of AI techniques, such as topic modelling and network analysis, in generating unprecedented research hypotheses and patterns; 2) deconstructing the fundamental epistemological limitations of automated analysis, notably its failure to comprehend historical context and perform source criticism; 3) proposing a hybrid integrative model that moves beyond the human-machine duality, fostering a synergistic interaction where the historian provides interpretative context while AI reveals latent quantitative insights; and 4) formulating ethical and practical guidelines to safeguard methodological rigor and mitigate algorithmic bias. The study concludes that the principal challenge resides not in the technical application, but in the critical assimilation of these tools into historical practice. It ultimately seeks to advance a balanced methodological vision, paving the way for a profound re-examination of the human textual heritage.