Analysing the Role of Generative AI in Software Engineering - Results from an MLR


Bazzan T., Olojo B., Majda P., Kelly T., YILMAZ M., Marks G., ...Daha Fazla

31st European Conference on Systems, Software and Services Process Improvement, EuroSPI 2024, Munich, Almanya, 4 - 06 Eylül 2024, cilt.2179 CCIS, ss.163-180 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 2179 CCIS
  • Doi Numarası: 10.1007/978-3-031-71139-8_11
  • Basıldığı Şehir: Munich
  • Basıldığı Ülke: Almanya
  • Sayfa Sayıları: ss.163-180
  • Anahtar Kelimeler: Benefits, Generative AI, Large Language Models, Risks, Software Engineering
  • Gazi Üniversitesi Adresli: Evet

Özet

Generative Artificial Intelligence (GenAI) has become a practical tool that exhibits the potential to revolutionize numerous industries through publicly available systems with simple yet effective interfaces. This paper outlines the findings of research conducted in a multivocal literature review (MLR) with the aim of exploring the impact of GenAI in software engineering, with a focus on the fundamental aspects, use cases, benefits, and risks associated with contemporary GenAI models leveraged in key industries and practices. Key findings indicate that GenAI is adopted in software engineering, with various reported benefits in areas including requirement engineering, estimation and testing. However, there are also some risks associated with GenAI-based Software Engineering, such as in the context of generated data consistency and accuracy (sometimes referred to as the Hallucination problem), plagiarism, bias, and security. GenAI-assisted software engineering is becoming more mainstream, but resolving all the associated issues is going to take some time.