Assessing AI Chatbots Performance in Comprehensive Standardized Test Preparation; A Case Study with GRE

Résumé : This research paper presents a comprehensive evaluation of the performance of three artificial 10 intelligence chatbots: Bing, ChatGPT, and GPT-4, in addressing standardized test questions. Graduate record examination, known as GRE, serves as a case study in this paper, encompassing both quantitative reasoning and verbal skills. A total of 137 quantitative reasoning questions, featuring diverse styles and 157 verbal questions categorized into varying levels of difficulty (easy, medium, and hard) were administered to assess the chatbots' capabilities. This paper provides a detailed examination of the results and their implications for the utilization of artificial intelligence in standardized test preparation by presenting the performance of each chatbot across various skills and styles tested in the exam. Additionally, this paper explores the proficiency of artificial intelligence in addressing image-based questions and illustrates the uncertainty level of each chatbot. The results reveal varying degrees of success across the chatbots, demonstrating the influence of model sophistication and training data. GPT-4 emerged as the most proficient, especially in complex language understanding tasks, highlighting the evolution of artificial intelligence in language comprehension and its ability to pass the exam with a high score.