Brief State of the Art on Human-AI in Software Engineering: Impact, Ethical Challenges, and Academic Evolution

Abstract

This article presents a brief state of the art on the integration of artificial intelligence (AI) in software engineering, focusing on its impact, emerging ethical challenges, and the academic evolution required to address these changes. Through a systematic literature review based on the PRISMA methodology and the analysis of case studies, it explores how AI, particularly large language models (LLMs), is transforming key processes such as code generation, program verification, and integration with DevOps practices, thereby improving software productivity and quality. However, these advances entail significant ethical challenges, such as the introduction of biases, lack of transparency in automated decisions, and excessive reliance on automation, raising concerns about privacy and security. This study highlights the importance of reformulating curricula, developing models for academic institutions that integrate an ethical approach to AI teaching, and fostering collaboration with the industry. This work offers a comprehensive vision that combines the technical, ethical, and educational aspects necessary for the responsible implementation of AI. Finally, directions for future research centered on developing more transparent AI technologies and creating robust ethical frameworks to ensure effective synergy between humans and AI in software engineering are identified.

Ähnliche Arbeiten