History Education and Generative AI: Navigating Accuracy and Bias

In the age of Artificial Intelligence (AI), the potential for AI-powered tools to revolutionize history education is immense, offering students worldwide quick and precise resources to investigate the depths of history. However, the critical factor lies in ensuring the accuracy of these tools to benefit learners truly. To address this concern, a qualitative research study was conducted to explore the research gap surrounding AI-generated responses related to two historical conflicts with conflicting historical narratives: Palestine-Israel and Ukraine-Russia. The aim was to scrutinize the accuracy and potential biases present in AI-generated content. The findings from the study revealed that ChatGPT, an AI language model, exhibited bias that reflected the political stance of the United States when responding to prompts related to historical conflicts. While the use of ChatGPT for historical learning purposes has shown promise in enhancing access to historical information, the study highlighted the model's inherent bias, indicating its unreliability when compared to primary historical sources. As a result, there is a clear need for future research endeavors to focus on developing inclusive and impartial AI-powered learning models to ensure accurate and unbiased historical education.