
How Do Developers Blog?
Our work on data-mining blogs and related artifacts in open source communities received the MSR'21 Most Influential Paper Award. Time to reflect.
Recently reported issues concerning the acceptance of Artificial Intelligence (AI) solutions after deployment, e.g. in the medical, automotive, or scientific domains, stress the importance of RE for designing and delivering Responsible AI systems. In this paper, we argue that RE should not only be carefully conducted but also tailored for Responsible AI. We outline related challenges for research and practice.
Over the last ~3 years, I have been thinking quite a bit about the importance of Requirements and Quality Engineering for AI. Together with Yen Dieu Pham and Larissa Chazette, we put our thoughts into a short paper, which appeared at IEEE Computer as part of a special issue on Software Engineering for Responsible AI (edited by Qinghua Lu, Liming Zhu, Jon Whittle, and Bret Michael).
The main message of the paper is simple: In order to ensure the reliability and acceptance of AI- and ML- based systems in practice, we first need to make sure that AI Engineers and Data Scientists know the fundamentals of Software and Requirements Engineering. Moreover, there are at least 6 major challenges related to Requirements and Quality Engineering, which we should address as a community of research and practice. These are:
You can find more in the preprint. Enjoy reading and feel free to share your feedback per Email.