The paper presents findings about the labor conditions on the work behind AI development, with two main goals. First, the aim is to investigate the current labor conditions in the AI industry and point out the lack of quality of outsourced jobs offered by digital labor platforms and Business Process Outsourcing companies, calling attention to the challenges faced by workers and the human costs of AI systems development. Second, it presents a global research project results on how not only to carry on international studies on the topic but also how to use an action-research approach to generate impact through public scoring companies and encourage the adoption of best practices by them.

The paper’s methods are based on the project’s action-research approach and methodological framework. The project scores companhies based on principles that address the major issues that define labour relations, namely, pay, conditions, contracts, management, and representation.

The paper presents the result of the assessment of cloudwork platforms conducted in 2023. In addition, it paper will examine labour relations in BPO companies in the AI supply chain using a case study of the company Sama, which is based in the United States and operates in many African countries.

Among the findings are problems such as non-payment situations, lack of policies to ensure minimum wage, significant rates of unpaid labour, poor measures to promote health and safety, problems in management practices and weak collective bargaining practices.

Artificial Intelligence (AI) has woven into a supply chain of capital, resources and human labour that has been neglected in debates about the social impact of this technology. While the literature on critical AI studies have focused on algorithmic bias and opacity, the global production line that fosters AI innovation have drawn little attention. Building on Tsing’s concept of supply chain capitalism, this paper offers a journey through mines, semiconductor manufacturers, data centres, technological firms, data labelling factories and e-waste dumps by illustrating the complex, diverse, opaque and global structure of the supply chain of AI. Then, the paper moves into illuminating a case study drawn from three months of fieldwork on data centres in Mexico, revealing that algorithmic harms go beyond code pitfalls. A close examination into the supply chain capitalism of AI reveals that other types of eco-political frictions are arising, particularly in the context of fundamental and environmental rights. This demands a broader and critical perspective on AI studies by considering the entire capitalist production line of its industry—from mineral extractivism to e-waste dumps—and its environmental and political consequences.

Despite the rise in commercial applications of LLMs, scholarship has neglected an in-depth appreciation of the free contribution of subjects communicative social action as the engine of training data production as a necessary moment in digital processes of valorisation. This issue was popular in the analyses of the post-operaist tradition of free labour, but have since been missing in examinations of more recent technological developments, specifically in what concerns AI. Although the work of Baudrillard is semi-frequently evoked in descriptive critical assessments of new technologies, there is little integration of Baudrillard's work in contemporary studies of AI. This paper aims at contributing in this direction, by showcasing the utility of Baudrillard’s concepts of simulation, subject function, masses, and the social, for an understanding of immaterial free labour in the context of Large Language Models (LLMs).

Drawing from the recent phenomena of the sale of Reddit communications content to OpenAI as training data, I propose the notion of digital common as the pre-trained collected and recorded data of actual human communication through digital systems. I propose the framework of the subject function as expounded by Baudrillard, in both its individual and collective aspects, as a necessary conjuncture to understand how commercial applications of conversational LLMs fit into the broader landscape of digital political economy. I suggest that the role played in this specific application derives from the appropriation of freely generated user-data as constituting the digital common and as carrying a specific conception of subjectivity as functional.

In a field outside Lund in southern Sweden, the world’s most powerful neutron source is nearly ready for action. This neutron source is located at the heart of the European Spallation Source (ESS), a Big Science facility that – when fully operational – will produce some of the largest quantities of digital data in today’s data-dominated world.

Starting in 2027, scientists from all over the world will travel to this experimental facility to test samples, hoping for cutting-edge discoveries. Reliable capture and management of experimental data is essential to ensure that scientific results have a solid foundation. However, the visibility of those working with data management and the institutional support for such activities at such facilities has historically been low.

This paper tells the story of a unique research journey following the people responsible for designing and implementing digital infrastructures for experimental data at the ESS as the demand for professional technical support grows in the scientific user community. Drawing on indepth interviews conducted over three years, I explore what the emergence of this new professional class means for validation of these experts’ skills and for the production of “raw” data in Big Science.