Generative Artificial Intelligence for Software Engineering: Outlook and Conclusions

Table of Links

• Abstract and Introduction
• Background
• Research Approach
• Research Agenda
• Outlook and Conclusions
• References

5. Outlook and Conclusions

The field of GenAI in SE is an important and fast-moving SE research area, currently presenting opportunities as well as open challenges. GenAI has the power to redefine the way software is designed, developed, and maintained. The innovative capabilities of AI-driven systems promise to streamline processes, boost efficiency, and open up new avenues for creative problem-solving. However, there are also common challenges in adopting GenAI and specific challenges for SE activities, i.e., requirements engineering, software implementation, quality assurance, and SE education. Our research agenda, outlined in this paper, stands as one of the pioneering initiatives to systematically reveal the state of research as of Oct 2023. Software Engineering is a large spectrum that includes technical activities but also user-centric, process-driven and managerial aspects. A comprehensive view should consider insights and methodologies from research about software project development, software project management, professional competencies, software engineering education, and software businesses. Therefore, collaboration with experts from these diverse disciplines is essential to harness the full potential of this research area.

In light of the evolving nature of GenAI research, significant work remains to establish GenAI in SE as a mature research area. Substantial theories must be developed to provide a solid theoretical underpinning for the practical applications. Technological advancements should be validated in a realistic project context. The adoption of GenAI in SE can also be combined with other active research areas, e.g., sustainability, trustworthy systems, and education.

We acknowledge that this research agenda might still miss some important SE aspects where GenAI can be applied. RQs are presented to illustrate significant concerns in each SE area and are not meant to be exhaustive. The research agenda is open to additions of new tracks, topics, and RQs by other researchers interested in the research area. Through this collective effort and commitment from a diverse community of researchers, we can strengthen the foundations of GenAI in SE and ensure its relevance in a rapidly changing technological landscape.

Declaration of Generative AI and AI-assisted technologies in the writing process During the preparation of this work, we used ChatGPT version 3.5 and Grammarly services to edit and improve our writing, i.e. rephrasing sentences and fixing typos and language mistakes. After using this tool/service, we reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.