<p>This paper presents an experimental case study on the application of GPT‑4 in the domain of object-oriented software development. The study aims to assess the effectiveness of GPT‑4 in generating functional and high-quality code, exploring the role of prompt engineering in guiding the model to achieve specific software development objectives. In the experimental case studies, GPT‑4 was tasked with progressively complex programming challenges. The first case, a&#xa0;Java-based student management system, demonstrated GPT-4’s capacity to generate structured, functional code in response to specific prompts. However, it also revealed the need for precise, context-rich prompts to create maintainable and high-quality software solutions. The second case, the “Snake” game, presented a&#xa0;more complex scenario. GPT‑4 initially produced general approaches and pseudocode, requiring iterative prompting and error analysis to develop a&#xa0;functional game. This case highlighted the limitations of GPT‑4 in handling complex programming tasks without detailed guidance. The results of these studies suggest that GPT‑4 can be a&#xa0;valuable tool in software development when combined with thoughtful prompting strategies and a&#xa0;deep understanding of software development principles. However, the necessity of human expertise in critical decision-making and complex design tasks is underscored. The limitations of GPT‑4 in automated testing and its current inability to independently handle complex software projects are also discussed. The paper concludes with a&#xa0;discussion on the integration of generative AI in software engineering methodologies and the ethical implications of biases observed in GPT-4’s responses. This study contributes to understanding the potential and limitations of generative AI in software development and highlights the ongoing need for human oversight and expertise in the field.</p>