Faith Nyasuguta
Artificial Intelligence (AI) is often celebrated as the technology of the future, poised to revolutionize industries, simplify complex tasks, and drive global innovation. However, beneath the surface lies a troubling reality: the human workforce powering these systems is being grossly exploited, particularly in low-wage economies like Kenya.
Humans Behind the Algorithms
The narrative that AI will replace human labor has persisted for years, but the truth is more complex. AI systems rely heavily on a global workforce known as “humans in the loop”, individuals who train AI algorithms by labeling, sorting, and sifting data.
This grunt work, essential for improving AI models used by tech giants like Meta, OpenAI, Microsoft, and Google, is often outsourced to countries with large, low-cost labor pools.
In Nairobi, workers like Naftali Wambalo, a mathematics graduate and father of two, spend hours labeling photos and videos to teach AI systems how to identify objects and human characteristics. Tasks range from tagging everyday items in images to categorizing facial features by race.
“You are teaching the machines how to think and act like humans,” Wambalo explains. This work ensures that autonomous vehicles can avoid pedestrians and that AI models can detect diseases, among other applications. However, while the work is vital, the conditions under which it is performed reveal a darker side of the AI boom.
The Illusion of Opportunity
Kenya, with an unemployment rate as high as 67% among its youth, is desperate for jobs. Tech companies have seized this opportunity, positioning AI jobs as pathways to a better future. President William Ruto has aggressively courted these firms, offering financial incentives and highlighting Kenya’s status as a “Silicon Savannah.” But civil rights activists like Nerima Wako-Ojiwa argue that these opportunities come at a steep cost.
“It’s like modern-day slavery,” says Wako-Ojiwa. “The jobs are advertised as transformative, but they are exploitative.” Contracts are often short-term, with some lasting only days or weeks. Workers are paid meager wages, around $2 per hour gross, a stark contrast to the $12.50 per hour tech companies reportedly pay outsourcing firms like SAMA, which hire the workers.
Sweatshops of the Digital Age
These outsourcing firms, acting as intermediaries, shield tech giants from direct association with the poor working conditions. Workers are hired on a project basis, leaving them without job security. Wako-Ojiwa likens the workspaces to “AI sweatshops,” where the tools are computers instead of sewing machines.
The financial disparity is glaring. While companies like Meta and OpenAI profit immensely, the workers enabling their innovations live paycheck to paycheck. “They say, ‘Because you are in Kenya, $2 is enough,’ but that idea has to end,” says Wambalo.
Mental and Emotional Toll
Beyond low wages, the nature of some tasks exacts a heavy emotional toll on workers. Wambalo recalls being assigned to sift through explicit and violent content, including images of abuse, suicides, and graphic violence, to train AI models to recognize and filter harmful material.
For many, the psychological impact has been devastating. Workers report experiencing flashbacks, isolation, and strained personal relationships. “I used to enjoy my marriage, but after seeing explicit content every day, I started hating intimacy,” Wambalo shares. Another worker, Fasica, admits, “I find it easier to cry than to speak.”
Although companies like SAMA claim to provide mental health support, workers argue it is insufficient. “We need trauma experts who understand what we’re going through,” says Wambalo.
Unfair Practices and Wage Theft
Even outside of emotionally taxing assignments, the challenges persist. Workers for companies like Scale AI, which operates the platform Remotasks, describe a pattern of wage theft. They allege that their accounts were arbitrarily closed before payday, denying them compensation for completed work.
Ephantus, a former Remotasks worker, recalls, “They would claim policy violations and refuse to pay us for the work we had done.” The lack of legal recourse leaves workers powerless, a vulnerability that companies exploit.
A Loophole in Labor Laws
Kenya’s outdated labor laws, which do not account for digital labor, exacerbate the problem. Efforts to push for better protections are met with resistance, as companies threaten to relocate to neighboring countries with even looser regulations.
“We’re trapped,” says Wako-Ojiwa. “The government is so desperate for jobs that it overlooks exploitation.” This desperation allows tech giants to impose conditions they would never consider in their home countries.
The Fight for Justice
Amidst these challenges, workers are beginning to push back. Wambalo and nearly 200 other digital workers have filed a lawsuit against SAMA and Meta, alleging “unreasonable working conditions” that caused psychiatric harm. The case highlights the broader need for accountability in the tech industry.
While companies like OpenAI and Meta claim to prioritize fair wages and safe working conditions, their reliance on opaque outsourcing arrangements complicates enforcement. Activists argue that tech giants must take direct responsibility for the welfare of their global workforce.
A Call for Change
The exploitation of Kenyan workers highlights a broader ethical dilemma in the AI industry. As AI continues to evolve, the human labor driving its progress must not be overlooked. Workers like Wambalo are not just cogs in a machine; they are the backbone of technological innovation.
“If these companies want to keep doing business here, they need to do it the right way,” says Wambalo. “Pay us what we deserve and treat us with dignity.”
The Kenyan government, too, must step up. Updating labor laws to protect digital workers and holding tech companies accountable is essential. As Wako-Ojiwa points out, “It’s not enough to create jobs. They must be good jobs.”
The promise of AI is immense, but its potential should not come at the expense of human dignity. As the global tech community looks to Kenya and similar countries for talent, it must also recognize its responsibility to foster ethical and equitable labor practices.
RELATED:
