Under-recognized work
Without training data, ML cannot take place. Much of this data comes from paid clickwork (also called “platform work” [170] or “microwork” [558]), unpaid crowdsourcing, and unpaid user behavior capture. Clickworkers, mainly in the global south, perform repetitive data-labeling tasks for use in the training of ML models [558]. The market value of such annotations “is projected to reach $13.7 billion by 2030” [228] and the annotation industry is widely reported to have little concern for workers’ rights. Besides welfare and rights, the invisibility of this contribution arguably contributes to a misunderstanding of AI capabilities.7
ENTITY
3 - Other
INTENT
3 - Other
TIMING
1 - Pre-deployment
Risk ID
mit874
Domain lineage
6. Socioeconomic and Environmental
6.2 > Increased inequality and decline in employment quality
Mitigation strategy
1. Establish and enforce robust, value-chain-wide human rights due diligence, requiring responsible contracting practices across all tiers of suppliers. Contracts must mandate fair compensation (e.g., living-income benchmarks), predictable task volumes, realistic deadlines, and embed shared responsibility among buyers, platforms, and model developers. 2. Implement comprehensive worker welfare and protection safeguards, including funded health and safety measures, specific protections and support for exposure to harmful content, accessible and effective grievance channels, and formal remediation mechanisms for identified harms. 3. Adopt and comply with international governance standards (e.g., those proposed by the ILO or Fairwork principles) that require digital labor platforms to provide minimum labor rights, legal clarity, and full transparency regarding algorithmic monitoring, data collection, and any decisions that impact a worker's pay or tenure.