AI (not) against AI

• Дата: 2025
• Автори: Nesterenko, Oleksandr, Yatsuk, Petro
• Формат: Стаття
• Мова: Українська
• Опубліковано: Kyiv National University of Construction and Architecture 2025
• Теми: information technologies; digital transformation; future; security; threats
• Онлайн доступ: https://es-journal.in.ua/article/view/351692
• Назва журналу: Environmental safety and natural resources

The aim of the research is to identify information, technological and methodological approaches to artificial intelligence developing in modern conditions of digital transformation, and in the long term. The objectives of the research are the following questions: a) to conduct a systematic analysis of the main aspects of AI development, in particular by the category of AI security; b) to identify the main time trends in AI improvement and ensuring AI security; c) to assess the current state of AI, development directions in this area and solving problems of ensuring AI security. Given the specifics of the research topic, tools in the form of modern AI means, such as ChatGPT, Claude, Copilot, Gemini, were used to conduct the research. As criteria for assessing the validity of conclusions, it is proposed to rely on the relevance and pertinence of the results of AI means' responses to various prompts. The main topic of inquiries is devoted to key trends that characterize the development of artificial intelligence, how AI will develop in the near future and until 2100, and what consequences this will have for humanity. The main stages of AI development have been identified in accordance with the above predictions. The main results show that the above-mentioned AI means primarily indicate an increase in attention to the ethical and legal aspects of using AI, as well as the growing AI integration into various areas of human activity. At the same time, it was noted that it is important to maintain a balance between using AI to enhance human capabilities and ensuring that people do not lose their intellectual skills. Particular attention is paid to the possible moment of singularity, when AI will begin to develop at an exponential rate, which will lead to radical changes in society and technology, and the consequences of this process are difficult to predict with certainty. Such systematic reviews are suitable for the formation of information meta-resources, which can be used by responsible persons as well as directly by specialists in various fields to support decision-making regarding the formation of AI security tools.