In the rapidly evolving fields of Artificial Intelligence (AI) and Information Technology (IT), the linguistic landscape continuously shifts, presenting unique challenges for bilingual lexicography of the growing scientific domains and subdomains. This research introduces the AI-driven Method of the Triangulation Approach (AMTA), a pioneering framework designed to bridge the gap between computational linguistics and traditional dictionary compilation. The proposed framework utilizes the present capabilities of advanced language models, specifically GPT and Gemini, to automate and enhance the creation of bilingual dictionary entries tailored for Arab learners and researchers in AI and IT. This study outlines the methodology behind AMTA for the AI-assisted design of dictionaries, emphasizing its integration of engineered prompts to guide AI models (particularly Large Language Models–LLMs) in generating linguistically accurate and contextually rich dictionary entries. Furthermore, when integrating a comprehensive array of linguistic features—including collocations and usage examples derived through authentic corpora—the entries would impart knowledge and support effective language usage. Relying on carefully engineered prompts in data and translation retrieval ensures that the AI models produce precise and contextually relevant content. Preliminary results reveal significant improvements in entry quality, demonstrating AMTA's potential to revolutionize specialized lexicography, particularly in growing fields such as IT. AMTA's broader implications suggest future lexicographical endeavors, emphasizing its versatility across different languages and domains. It involves designing bilingual dictionaries to expand domains and subdomains and further maintain their contemporaneousness.