# The Data-Production Dispositif

arXiv: https://arxiv.org/abs/2205.11963 · HTML: https://mlsanity.com/paper/2205.11963/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 101 references that are in the map.
- First posted on arXiv 24 May 2022.

## Builds on (most cited first, 9 of 9)

- [Are We Modeling the Task or the Annotator? An Investigation of Annotator Bias in Natural Language Understanding Datasets](https://mlsanity.com/paper/1908.07898/) (2019)
- [Do Datasets Have Politics? Disciplinary Values in Computer Vision Dataset Development](https://mlsanity.com/paper/2108.04308/) (2021)
- [Garbage In, Garbage Out? Do Machine Learning Application Papers in Social Computing Report Where Human-Labeled Training Data Comes From?](https://mlsanity.com/paper/1912.08320/) (2019)
- [A Data-Driven Analysis of Workers' Earnings on Amazon Mechanical Turk](https://mlsanity.com/paper/1712.05796/) (2017)
- [Between Subjectivity and Imposition: Power Dynamics in Data Annotation for Computer Vision](https://mlsanity.com/paper/2007.14886/) (2020)
- [Data Vision: Learning to See Through Algorithmic Abstraction](https://mlsanity.com/paper/2002.03387/) (2020)
- [Measuring Social Biases of Crowd Workers using Counterfactual Queries](https://mlsanity.com/paper/2004.02028/) (2020)
- [The Future of Work Is Here: Toward a Comprehensive Approach to Artificial Intelligence and Labour](https://mlsanity.com/paper/2007.05843/) (2020)
- [We Haven't Gone Paperless Yet: Why the Printing Press Can Help Us Understand Data and AI](https://mlsanity.com/paper/2104.12731/) (2021)

## Built on by (most cited first, 16 of 16)

- [The Participatory Turn in AI Design: Theoretical Foundations and the Current State of Practice](https://mlsanity.com/paper/2310.00907/) (2023)
- [Participation in the age of foundation models](https://mlsanity.com/paper/2405.19479/) (2024)
- ['Simulacrum of Stories': Examining Large Language Models as Qualitative Research Participants](https://mlsanity.com/paper/2409.19430/) (2024)
- [Weakly Supervised Detection of Hallucinations in LLM Activations](https://mlsanity.com/paper/2312.02798/) (2023)
- [Using Text-to-Image Generation for Architectural Design Ideation](https://mlsanity.com/paper/2304.10182/) (2023)
- [Perceptions of Sentient AI and Other Digital Minds: Evidence from the AI, Morality, and Sentience (AIMS) Survey](https://mlsanity.com/paper/2407.08867/) (2024)
- [Ground Truth Or Dare: Factors Affecting The Creation Of Medical Datasets For Training AI](https://mlsanity.com/paper/2309.12327/) (2023)
- [AI Alignment through Reinforcement Learning from Human Feedback? Contradictions and Limitations](https://mlsanity.com/paper/2406.18346/) (2024)
- [The Necessity of AI Audit Standards Boards](https://mlsanity.com/paper/2404.13060/) (2024)
- [Quality Assured: Rethinking Annotation Strategies in Imaging AI](https://mlsanity.com/paper/2407.17596/) (2024)
- [Building Benchmarks from the Ground Up: Community-Centered Evaluation of LLMs in Healthcare Chatbot Settings](https://mlsanity.com/paper/2509.24506/) (2025)
- [The Ethics of AI Value Chains](https://mlsanity.com/paper/2307.16787/) (2023)
- [A toolkit of dilemmas: Beyond debiasing and fairness formulas for responsible AI/ML](https://mlsanity.com/paper/2303.01930/) (2023)
- [Making Data Work Count](https://mlsanity.com/paper/2311.18046/) (2023)
- [The Consensus Trap: Dissecting Subjectivity and the "Ground Truth" Illusion in Data Annotation](https://mlsanity.com/paper/2602.11318/) (2026)
- [Investigating Labeler Bias in Face Annotation for Machine Learning](https://mlsanity.com/paper/2301.09902/) (2023)

## Abstract (arXiv)

Machine learning (ML) depends on data to train and verify models. Very often, organizations outsource processes related to data work (i.e., generating and annotating data and evaluating outputs) through business process outsourcing (BPO) companies and crowdsourcing platforms. This paper investigates outsourced ML data work in Latin America by studying three platforms in Venezuela and a BPO in Argentina. We lean on the Foucauldian notion of dispositif to define the data-production dispositif as an ensemble of discourses, actions, and objects strategically disposed to (re)produce power/knowledge relations in data and labor. Our dispositif analysis comprises the examination of 210 data work instruction documents, 55 interviews with data workers, managers, and requesters, and participant observation. Our findings show that discourses encoded in instructions reproduce and normalize the worldviews of requesters. Precarious working conditions and economic dependency alienate workers, making them obedient to instructions. Furthermore, discourses and social contexts materialize in artifacts, such as interfaces and performance metrics, limiting workers' agency and normalizing specific ways of interpreting data. We conclude by stressing the importance of counteracting the data-production dispositif by fighting alienation and precarization, and empowering data workers to become assets in the quest for high-quality data.