What is it about?
In this study we interviewed expert stakeholders working in various positions involving AI technologies and asked them about AI governance in the system development life cycle. From the data we identified all unique aspects of AI governance and ultimately mapped them to three key parts of the DevOps lifecycle: (1) design; (2) development; and (3) operation.
Featured Image
Photo by DeepMind on Unsplash
Why is it important?
Our work can help stakeholders obtain a general understanding of which AI governance issues they need to consider and at which stages of the system development life cycle.
Perspectives
We wrote this article as part of a Business-Finland funded project called AI governance and auditing (AIGA). I would consider this article as a rough overview of the AI governance issues that stakeholders face, but more detailed and granual approaches are still needed for more holistic and mature AI governance frameworks. I was personally most satisfied with the quality of the interviewed experts and the collaboration between the authors where everyone helped and contributed.
Samuli Laato
Turun Yliopisto
Read the Original
This page is a summary of: AI governance in the system development life cycle, May 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3522664.3528598.
You can read the full text:
Contributors
The following have contributed to this page
