SS4:Generative AI and the Work of the User         

·   Organizers

Hao Wang, Xidian University, Xi'an, China - wanghao@xidian.edu.cn, Yiming Luo, Xidian University, Xi'an, China - 18737611666@163.com, Di Wu, Norwegian University of Science and Technology, Norway- di.wu@ntnu.no, Yushan Pan, Xi'an Jiaotong-Liverpool University, Suzhou, China - yushanp@liverpool.ac.uk, Yihong Wang, Xi'an Jiaotong-Liverpool University, Suzhou, China - Yihong.wang@xjtlu.edu.cn, and Zuhe Li, Zhengzhou University of Light Industry, Zhengzhou, China - zuheli@zzuli.edu.cn,

·   Description/Abstract

We invite submissions for a special edition of the CSCWD conference titled "Generative AI and the Work of the User." The CSCW community has a longstanding interest in and commitment to sociotechnical perspectives. In line with this community's interests, our focus is on understanding the work that surrounds the technical aspects of developing, using, and refining generative AI as a tool. Various forms of artificial intelligence, including supervised, unsupervised, reinforcement learning, or their combinations, all rely on decision-making by human actors in some capacity. The same applies to the users who utilize the outcomes of these AI systems. Consequently, generative AI also delves into the age-old discussion of the relationships between rules and regularities, as well as between correlation and causation.

We acknowledge the distinction between generative AI and artificial intelligence (AI), although they are closely related concepts. Generative AI specifically refers to computer programs that, through the use of algorithms, aim to produce improved and more accurate results. Whether these results are indeed more accurate and useful depends on several investigable factors. Additionally, significant efforts are devoted to data classification and data input in generative contexts. Understanding both the input and output of generative AI tools is a complex and interpretable matter. Furthermore, the deployment of generative AI raises serious ethical issues, which can be highlighted or obscured in practical use cases.

In summary, the 'work to make generative AI work' can be extensive and requires skill. Research studies that shed light on these issues might encompass:

• Research on the creation of initial datasets for generative AI, especially in the context of Metaverse applications (XR, AR, VR) across consumer and industrial sectors.
• Research on data interpretation and analysis, exploring various contexts including knowledge graph utilization and data mining in both academic and industrial contributions.
• Research into the adoption of generative AI results by expert and non-expert users in practical domains (non-experimental), e.g., Software Development.
• Exploring specific usage contexts, including healthcare, law, bioinformatics, consumer behavior analysis, behavior profiling, individual rights, sentiment analysis, generative AI-driven choice architectures, software engineering, resource management, and optimization.
• Exploring the potential for 'explainable' or 'explorable' generative AI in collaborative work settings, such as design workshop and visual arts.
• Theoretical approaches like critical studies on the trustfulness of the generative AI and their application in contexts.
• Addressing the political and ethical concerns arising from reliance on generative AI results, including issues related to trust.
• Exploring the epistemological implications of generative AI-driven analysis beyond the listed contexts.