Intelligenza artificiale generativa e memoria storica: la problematica del revisionismo tra condiscendenza e allucinazioni

Abstract

This essay examines the role of generative artificial intelligence (AI), such as ChatGPT and Gemini, in shaping historical memory and the risks of ideological revisionism. Historical memory, as a collective cultural construct, is vulnerable to manipulation, especially through AI’s tendencies toward compliance and “hallucinations”. The study distinguishes between critical revisionism, which advances scholarly research, and ideological revisionism, which distorts history for political or social purposes. Through a case study on the Ardeatine Caves massacre, the author evaluates how generative AI models handle historical inaccuracies and falsifications, revealing varying resistance to manipulation. While generative AI offers unprecedented opportunities for historical research, its potential to legitimize revisionist narratives raises ethical concerns.