The use of corrective feedback from generative artificial intelligence in teaching a professional foreign language to students of an agricultural university

Importance. The methodological potential of evaluative corrective feedback from the generative artificial intelligence (AI) means is beginning to be used by teachers in teaching learners and students written foreign language utterance. At the same time, the use of extracurricular practice by students of a non-linguistic university with tools and in order to receive corrective feedback in the subject-language integrated teaching of a professional foreign language has not been studied separately. The goal of the work is to develop the methodology stages for using corrective feedback from generative AI in teaching a professional foreign language, conducting experimental training and empirically verifying the effectiveness of this technique. Materials and Methods. The study involved students of the Veterinary Medicine department of Voronezh State Agrarian University named after Emperor Peter the Great. The students of the control group ( N = 43) participated in subject-language integrated learning without using generative AI tools. The students of the experimental group ( N = 43) participated once a week in extracurricular work with the DeepSeek neural network in order to receive evaluative corrective feedback when performing integrated tasks. During the experiment, three aspects were controlled: a) the lexical side of speech; b) the grammatical side of speech; c) the professional content of the utterance. The Student’s t -test is used for statistical analysis of the data. Results and Discussion. The study proved the methodology effectiveness of using evaluative corrective feedback from generative AI in subject-language integrated learning in all controlled aspects: a) lexis ( t = 5.24 at p < 0.05); b) grammar ( t = 4.74 at p < 0.05); c) the professional content of the utterance ( t = 6.04 at p < 0.05). Conclusion. In the course of the study, a step-by-step methodology is developed for using evaluative corrective feedback from generative AI in subject-language integrated learning. The perspective of this study lies in using an approach to integrate students’ practice with professionally oriented AI tools into the subject-language education of students at a non-linguistic university.