Worker Exploitation Rife in AI Industry

TL/DR:

Millions of workers are stuck in low-paying, dead-end jobs in the AI supply chain. Despite the industry's wealth, there is little upward mobility or fair pay for these workers. The exploitation of hidden workers in the AI industry is a growing concern.

The AI industry is booming, with generous funding and highly paid executives and researchers at the top. However, the supply chain churning out generative AI tools like ChatGPT has a darker side. At the bottom of the chain are millions of workers who toil at screens, training algorithms for meagre wages.

According to a recent World Bank estimate, between 150 million and 430 million people do such work. They annotate images, text, and audio, create bounding boxes around objects in images, and even write haikus, essays, and fictional stories to train the sophisticated tools that could eventually replace them.

The Hidden Workers of the AI Industry

These workers are the hidden backbone of the AI industry, but they are often just scraping by. There is precious little of the time-honoured aspiration for the developing world: upward mobility. Data workers are still confined to low-value work, with few opportunities to transition to higher-paying digital jobs. This echoes broader discussions about What Every Worker Needs to Answer: What Is Your Non-Machine Premium? in an AI-driven economy.

Leaders of data-labelling firms often start with noble intentions to help pull people out of poverty. However, they have struggled to get corporate customers to pay higher rates as competition in their field has increased. As a result, most data work platforms do not have policies in place to ensure their workers earn at least the local minimum wage. This situation contributes to the broader concern of a trust deficit in AI in Southeast Asia.

Exploitation of Hidden Workers

The exploitation of hidden workers in the AI industry is a growing concern in Asia. For example, a recent job ad in Nigeria sought "professional translators" in Igbo for up to $17 an hour to help train generative AI models. This is well below the average rate for Nigerian translators, who tend to start at $25 an hour.

The ad came from Remotasks, the main platform of San Francisco-based AI startup Scale.ai, which just raised $1 billion from investors including Amazon.com in one of the year's largest financing rounds. Scale.ai and its rivals, such as San Francisco-based Samasource Impact Sourcing, Argentina's Arbusta S.R.L., and Bulgaria's Humans in the Loop, play a critical role in the AI supply chain. However, they typically pay just enough for workers to maintain a living. This issue of fair compensation is central to the concept of ProSocial AI.

The Future of Hidden Workers in the AI Industry

As data work becomes more complex, platforms like Scale.ai have been looking for more skilled workers, including artists and people with creative-writing degrees to write short stories for training AI systems. While these jobs offer higher wages, they are still below what people with degrees should be earning.

Researchers say the appetite for such work is growing, but with few incentives to provide an equitable wage, it is hard to see workers' economic status improving. The AI industry's true transformative effects have been in entrenching economic power, with fewer opportunities for the people underpinning the AI revolution.

Towards Fair Pay for Data Workers

Perhaps we can learn something from Nike's experience in the 1990s. The company faced an enormous backlash for the long hours and meagre wages its workers in developing nations earned. Over time, consumer boycotts and pressure from the media led Nike to put in stricter labour policies. It spent millions of dollars on improving conditions and pay. A detailed analysis of the ethical implications of AI development and deployment can be found in a report by the AI Now Institute entitled The AI Now Institute 2019 Report[^1].

The challenge for data workers is that their jobs are harder to visualise in the same, concrete way you can imagine a young boy sewing tennis shoes in a dimly-lit warehouse. However, tech companies should remember that poor working conditions at the bottom of their supply chain can also lead to substandard AI, impacting the quality of tools like those mentioned in Free Chinese AI claims to beat GPT-5.

Comment and Share:

What do you think about the exploitation of hidden workers in the AI industry? How can we ensure fair pay and opportunities for upward mobility for these workers? Subscribe to our newsletter for updates on AI and AGI developments, and share your thoughts in the comments section below.

[^1]: The AI Now Institute 2019 Report

YOUR TAKE

We cover the story. You tell us what it means on the ground.