We invite abstracts and proposals for an
international workshop on Data Work
Bengaluru, India (March 2025)
hosted by The University of Amsterdam, Aapti Institute, and Tattle
We invite contributions to a two-day workshop that explores ‘data work’—tasks such as data labelling and annotation that are necessary for the creation and deployment of artificial intelligence (AI). This workshop aims to foreground the hidden human costs and epistemic questions involved in the development of datasets and AI systems, which often obscure the vast amounts of data and labour required for their construction and maintenance.
Popular narratives around AI often focus on its potential to surpass human intelligence and transform professional domains and social contexts. AI development has received heightened attention, entrepreneurial interest and investment in India, amidst ambitions to pioneer advancements in language model development. As efforts to scale, adapt and replicate AI models soar, fuelled by hype around AI, the crucial role of human input to train, fine-tune, evaluate, and moderate these systems remains largely obscured. This is despite India being a globally prominent site for data work engaging millions of workers in crowdsourced and outsourced data work. Much of this work entails cognitively demanding tasks, involving datasets from varied (and sometimes, fraught) social contexts, and often performed under difficult conditions of work.
This movement of data, labour, resources and innovation within/along the globalised dataset production chain for AI also reflects the extractive legacies of colonial value chains. Data-driven and automated systems are constituted as ideal, objective and universal, while the situated knowledge, skill and care involved in data work that scaffolds the entire enterprise is invisibilized. This raises important epistemic questions regarding what forms of knowledge and expertise are considered valid and valued, and consequently, what forms of knowledge might be discounted or delegitimised.
In this workshop, we aim to foreground the contributions and concerns arising from this oft overlooked aspect of AI production and examine the colonial legacies, epistemic concerns and labour implications of data work in India and broader the Global South.
Day 1 of the workshop will feature interactive, hands-on activities and exercises designed to set the stage for the discussions on Day 2. These may include activities such as mapping exercises, data work simulations and walkthroughs.
Building from the Day 1 activities, we invite contributions from participants for Day 2 of the workshop that explore the global value chains of labour and knowledge that make up AI systems. Submissions may take various forms, and we encourage both academic and non-academic approaches. Possible contributions include, but are not limited to:
- Academic Research: Theoretical or empirical work that explores the material and epistemic dimensions of data work.
- Work-in-Progress Proposals or Pitches: Early-stage projects or ideas related to the human, social, and epistemic aspects of data work in AI development.
- Policy or Advocacy Proposals: Contributions that explore ways to address the unregulated or extractive conditions of data work and suggest frameworks for improving workers’ rights and wellbeing.
- Practice-based reflections: Personal or organisational accounts that draw from lived experience in the domain of data work or AI development.
- Creative or Artistic Contributions: Artistic works that creatively engage with the theme of data work and the hidden labour behind AI systems.
- Expressions of interest: If you do not have any tangible work to contribute, you can still submit a motivated expression of interest.
Provocations to Explore
Submissions can, but need not, relate to some of the following provocations:
- To what extent does AI development depend on the availability of labour that is largely a product of colonial legacies? How does it impact how such work and skill is valued?
- Where do data workers fit in India’s national and international ambitions with respect to technology and artificial intelligence?
- How can we remain reflexive about the extractive dynamics within AI development and research?
- Can data work become a site for epistemic plurality, rather than dominance?
- Can AI supply chains be made fairer, more transparent and more accountable?
Important Dates
- Submission of Abstracts / Expression of Interest: January 24, 2025
- Notification of acceptance: January 31, 2025
- Workshop: Two days in the week starting March 17, 2025
Submission Information
We invite a short abstract (250–300 words) describing your proposed contribution or expression of interest. You may also include supplementary materials such as links, images, videos, or prior work that supports your submission. We aim to cover accommodation and travel expenses within India, with the possibility of limited support for international travel.
The abstract should be submitted via email to MappingDataWork@proton.me Make sure to include your name and institutional affiliation (if any).
Please also indicate whether you will require support for travel and accommodation.
Practical Information
In case of any questions, please reach out to MappingDataWork@proton.me
Organising Committee
Aditya Singh (University of Edinburgh)
Jef Ausloos (University of Amsterdam)
Srravya Chandhiramowuli (University of Edinburgh)
Valentina Carraro (University of Amsterdam)
Aapti Institute
Tattle