Introduction

At present, interest in the application of artificial intelligence in the educational sphere is steadily increasing. Despite this, scientific studies specifically devoted to teaching English through smart technologies are still relatively limited. Existing literature on the topic mainly examines a variety of machine-intelligence-based applications within foreign language learning. A number of researchers, including P.V. Sysoev, E.M. Filatov, and Yu.E. Valkova, have contributed significantly to the development of this field. International studies assess the use of GPT-based chat systems for writing and editing academic works, emphasizing their role in reducing the time and effort required from both teachers and learners and in generating ideas while improving the quality of written texts.

Meanwhile, local scholars focus on the influence of GPT chat on students’ cognitive activity and on the integration of smart technologies into education, identifying their strengths and weaknesses. There are also viewpoints that highlight the limitations of the system and question the appropriateness of using GPT chat as a co-author of scientific publications. In addition, certain experts analyze the effects and outcomes of student learning when interacting with chat-based technologies. Nevertheless, exploring the potential of GPT chat remains timely, as the system continues to evolve and produces different results with changing modes of interaction.

Thus, incorporating digital technology into education has become a necessity and should form an essential component of the pedagogical process for preparing competitive specialists in any field. In this regard, it is crucial to carry out scientific research aimed at testing and analyzing modern digital systems to determine the optimal ways of integrating smart technologies into education – including foreign language learning, specifically English, in higher education. Such inquiry will help identify both the capabilities and limitations of these systems.[1]

The scientific novelty of the research lies in the potential use of chat agents as part of the English-teaching process in technical higher educational institutions, in the context of rapid digitalization and the modernization of teaching practices. The research investigates the practical application of GPT chat for reviewing students’ written assignments and identifies errors in its autonomous operation, using GPT chat as a supportive tool. The purpose of the study is to examine whether GPT chat can be effectively used for automated assessment of students’ written work in English courses at a technical university.

The research objectives include: establishing practical collaboration with a chatbot to determine whether it is feasible to use GPT chat for evaluating written assignments, analyzing the system’s capabilities, and identifying the efficiency and limitations of the digital assistant in the process of automatic assessment. The theoretical significance of the study lies in broadening the understanding of how digital technologies – including artificial intelligence – can support teachers in teaching English in higher education. The practical significance, in turn, is defined by the possibility of using GPT chat in teaching practice to reduce routine workload and shorten the time teachers spend assessing written tasks.

Discussion and results

Modern technologies and digitalization are evolving at an unprecedented pace and are being applied across almost every sphere of life. In the past, technologies were mainly used in professional environments and seemed unfamiliar to ordinary people. However, technological progress continues to advance, and today cutting-edge systems are widely utilized in various fields. One of the most significant and rapidly developing tools of modern technology is artificial intelligence. Although there are numerous definitions of the term «Artificial Intelligence,” they all describe software systems designed to imitate human cognitive processes. The main characteristic that distinguishes artificial intelligence from traditional technological systems is its ability to self-learn and improve while performing tasks, allowing it to adapt to constantly changing demands.[2]

The twenty-first century can undoubtedly be referred to as the era of innovation. Telecommunications technologies, which play a crucial role in today’s fast progress, have recently stimulated the massive expansion of artificial intelligence into different areas of human activity. Whereas in the past AI was primarily restricted to industrial production, robotics, and digital systems performing complex calculations, it now plays a significant role in various social domains – particularly in the field of education and, more specifically, in the teaching and learning of the English language. The rapid development of digital tools has had a strong impact on the educational process, as the traditional education model does not always meet the needs and expectations of today’s learners. Therefore, the learning environment must be adapted to modern realities, with intelligent technologies replacing outdated methods and making the educational process more dynamic and efficient.

The integration of artificial intelligence into education is commonly classified into three dimensions: support for students, support for teachers, and the comprehensive digital transformation of the learning process.

Among the AI tools most widely used in teaching English today are systems capable of identifying and analyzing text – including voice assistants, chatbots, online translation services, as well as platforms designed to check spelling, punctuation, grammar, and stylistics.[3]

This research focuses on the potential of intelligent technologies as a support tool for teachers, with particular attention to the use of the GPT-3.5 artificial intelligence-based chatbot in assessing students’ academic skills.[4]

Recently, OpenAI has introduced its newest products – the GPT-3.5 and GPT-4 chatbots – which have greatly simplified the integration of artificial intelligence into education and demonstrated positive results in improving learning efficiency. The updated models represent major advancements in the field of text generation tools and applications. Compared to previous systems, they show a remarkably high capability for producing written content, especially long-format essays and creative pieces, and excel in performing human-like actions across various professional and academic tasks. These developments mark a significant breakthrough in the field of text generation. Contemporary academic discussions increasingly highlight the potential role of GPT-based chat systems in solving various writing-related tasks.

This study is devoted to evaluating the relevance and efficiency of using GPT-3.5 and GPT-4 chatbots as tools for assessing students’ written assignments in the process of teaching English at the university level. Today, artificial intelligence-based programs such as ChatGPT, Yandex AI, Google AI and similar platforms are actively used by school learners, university students, postgraduate researchers, and independent scholars in their everyday academic tasks. Considering that most learners become familiar with artificial intelligence technologies already during their school years, the issue is not whether to restrict their use, but rather how to ensure their reasonable and purposeful application. This highlights the need to regulate and monitor their use and to cultivate a culture of effective and responsible application of AI tools both inside and outside the classroom.[5]

The development of artificial intelligence largely focuses on the creation and representation of knowledge-based systems. This direction deals with constructing knowledge models and knowledge bases, which form the foundation of expert systems. For example, AI-based systems can be used to test students’ knowledge of English, identify their current proficiency level, diagnose their challenges, and classify learners into differentiated groups for targeted teaching. In this way, English learners may be divided into:

1. a) beginner or low-level learners,
2. b) intermediate-level learners,
3. c) high-level or advanced learners.[6]

Conclusion

Artificial intelligence technologies are introducing a qualitatively new approach to the language teaching process. They serve to establish individual learning, direct the student to work on errors, increase educational efficiency and facilitate the teacher's work. At the same time, the full integration of AI capabilities into the educational process requires technical infrastructure, methodological updating and specialist training. The rational implementation of these technologies creates a solid foundation for further improving the quality of language teaching and the development of a digital education system.