SPECIFIC FEATURES OF THE USE OF ARTIFICIAL INTELLIGENCE IN MEDIATION AND FORECASTING OF THE NEGOTIATION PROCESS IN INTERNATIONAL ECONOMIC RELATIONS

Abstract

The purpose of the study is to determine and systematize the role of artificial intelligence in mediation and forecasting the negotiation process in international economic relations, as well as to substantiate the possibilities and limitations of its practical application. The methods of theoretical generalization, systematic and comparative analysis of modern scientific research in the field of mediation, negotiations and artificial intelligence were applied; content analysis of scientific publications, cases and digital tools (CalendarHero, Sonix, Descrybe.AI, AI models); logical-structural approach to building a matrix of possibilities and limitations of AI in negotiations and forecasting. It was established that the integration of AI transforms mediation from a predominantly intuitive and communicative practice into a more analytically guided process, where algorithms perform the functions of data processing, information structuring and decision-making support. It is shown that AI increases the efficiency of mediation through the automation of administrative tasks (scheduling meetings, transcribing, summarizing), improving access to legal information and analytical support for the mediator. It is substantiated that the use of AI models allows generating draft mediation agreements, modeling dialogue scenarios and supporting the training of mediators. It is proven that AI is able to increase the objectivity and consistency of the mediation process, but cannot fully replace human empathy, ethical assessment and creativity in forming compromises. In the field of forecasting the negotiation process, it is established that AI performs the function of a scenario modeling and early warning tool, allowing to assess the probability of concluding an agreement, the risks of escalation, the potential limits of compromise (ZOPA) and the possible consequences of individual negotiation steps. It is shown that forecasting is based on the analysis of historical negotiation data, text communication and contextual variables (economic conditions, deadlines, alternatives of the parties). It was found that the greatest predictive accuracy is achieved in formalized B2B negotiations, where the parameters of the agreement are clearly defined. The results of the study are useful for mediators, negotiators in the field of international economic relations, lawyers, analysts, developers of digital platforms for dispute resolution, as well as for scholars who study the digital transformation of negotiation processes.