Who’s powering AI? Shining a light on the AI data work industry.

AI tools have entered almost all aspects of our daily lives—from ordering groceries to writing emails, and even picking presents for and checking in on friends and family. While there is a dawning realization that our data fuels these tools, few are aware of the thousands of data workers in the US and around the world who build and maintain AI. 

Who are data workers?

Between the frontend of AI applications that we see on our phones and computers and the backend of coders and engineers who program large language models (LLMs), there are long chains of workers who generate and annotate the data that feeds AI systems, train and test these systems, and even fill-in for them when they fail to perform as intended. 

These data workers work for companies in an industry that is highly fragmented and opaque. Starting from Big Tech companies in Silicon Valley and elsewhere, chains of subcontracting relationships stretch across the world. Global subcontracting has long been a common business strategy for companies trying to increase margins by cutting labor and compliance costs by outsourcing work domestically or offshoring work to countries with poorly enforced labor laws. Companies and workers who contribute different knowledge and resources to AI systems are also part of a global value chain in which they co-create these powerful tools. 

Workers themselves often don’t know to which AI systems or tools they’re contributing. Data work tends to be low-wage and precarious, and workers report mental health impacts from certain aspects of this work. Data workers struggle to make ends meet even as companies raise billions of dollars in private equity and venture capital. 

Pulling back the curtain

For workers and their advocates to improve data work, the first challenge is understanding which companies operate in this space, what they do, and how they are related to each other. The current lack of transparency across the industry makes it difficult to understand where power and influence lies among different companies in the AI value chain and how those companies influence working conditions. The global and fragmented nature of the work, which takes place both in physical offices and on platforms in workers’ homes, and the constant shifting of companies’ contracts, also poses a challenge for workers and organizations seeking to find workers to be able to bring them together to fight for improved conditions.

We created the Data Work Landscape to help address this challenge. By highlighting some of the companies that employ data workers, we can begin to throw light on this shadowy industry—and provide a resource to workers building power in this industry. 

Delve into more resources

The landscape provides a snapshot of some of the companies that employ data workers, but this resource is not – and cannot be – complete. This project is one among many that are unveiling the data work industry and the workers who power it. Researchers, workers, and advocates worldwide are filling in the gaps collectively, and this work wouldn’t be possible without everyone’s contributions. If you would like to learn more about the global data work portion of the AI value chain, along with how labor contributes in many ways to the creation of digital technologies, here are additional resources that have informed the TechEquity team’s understanding of these issues: