Identifying Ambiguity Using AI: Insights from German, English, and Macedonian Language Students at the Faculty of Philology, Goce Delcev University, Stip

This study investigates the application of Artificial Intelligence (AI) in identifying linguistic ambiguity, with a focus on German-, English-, and Macedonian-language texts. Ambiguity—lexical, syntactic, referential, and pragmatic—represents a key challenge for clarity and may lead to misinterpretation in both everyday and academic communication. The research examines how AI-based language tools, particularly ChatGPT, can support students in detecting ambiguous expressions and exploring alternative interpretations.
The study involved approximately 30 students from the Faculty of Philology at Goce Delcev University – Stip, divided into three groups: students of German Language and Literature, English Language and Literature, and students of Macedonian Language or those studying German/English as elective foreign languages. Each group was given tasks in their target language containing ambiguous expressions and was instructed to identify and analyze them using AI tools.
Findings indicate that AI significantly expands the range of detected meanings compared to students’ initial interpretations. Students typically identify dominant and prototypical meanings, whereas AI generates a broader set of interpretations, including less frequent or context-dependent ones. However, students demonstrate stronger accuracy in evaluating contextual appropriateness, particularly in cases involving pragmatic and culturally bound meanings.
The study highlights the pedagogical potential of AI language models such as ChatGPT, not only as tools for ambiguity identification, but also as instruments for developing linguistic awareness, critical thinking, and advanced academic writing skills in multilingual contexts.