DIALOGUE INTERACTION WITH AI AS A TOOL FOR DEVELOPING STUDENTS' SCIENTIFIC SUBJECTIVITY

Abstract

Summary. The rapid integration of Large Language Models (LLMs) into higher education has created significant challenges for academic integrity and traditional methods of evaluating students' research work. There is a critical risk of "cognitive offloading," where students delegate intellectual labor to algorithms, subsequently losing their research subjectivity and agency. The problem lies in the lack of methodological tools that would allow the integration of AI into the educational process not as a substitute for thinking, but as an "intellectual sparring partner" that stimulates critical analysis and deep inquiry.

The purpose of the article is to substantiate and develop a methodical model of dialogue interaction with AI based on the principles of "digital maieutics" to foster student scientific subjectivity and ensure transparency in the research process.

Methods. The study employs a systematic approach to analyze "human-AI" interaction. The method of digital maieutics (an adapted Socratic dialogue) is applied to stimulate cognitive search. The concept of "Hamilton’s Risk" is used to differentiate between AI hallucinations and genuine scientific gaps. The methodology relies on iterative cycles of data verification through authoritative scientometric databases (Scopus, Web of Science) and a reflexive audit of prompt engineering logs.

The authors propose the "Semantic Resonance" model, where AI acts as a mediator helping students identify lacunae in their own knowledge. A multi-level verification protocol has been developed, including mandatory AI-Disclosure. It is proven that assessment should shift from the final text to the analysis of the research's "Digital DNA" — interaction logs that demonstrate the evolution of the student's thought. The "Aquarium" format for defending results is introduced, where students publicly verify AI-generated theses, proving their own expertise and authorship.

For the first time, the concept of "digital maieutics" is integrated into the structure of preparing scientific papers, allowing the legalization of AI use while preserving the author's scientific agency. An approach to identifying a student's personal intellectual contribution through an audit of their dialogue trajectory with AI is formulated. The role of "Hamilton’s Risk" is substantiated as a catalyst for seeking scientific novelty in areas where AI tends to hallucinate.

Transitioning from output control to process audit is the only way to maintain academic quality in the AI era. The authors propose implementing mandatory modules on critical prompt engineering and data verification into the curriculum. It is recommended to use the developed protocols of dialogue interaction as part of the reporting for students' research internships, ensuring the development of their subjectivity and adherence to the principles of academic integrity.