'AI Is African Intelligence': The Workers Who Train AI Are Fighting Back
In recent years, the rise of artificial intelligence (AI) has brought to light the often-overlooked human labor that fuels its development. In Kenya, workers like Michael Geoffrey Asia have become the backbone of AI training, moderation, and even intimate interactions through sex chatbots. Despite their critical role, these workers face significant challenges, including low pay, mental health issues, and a lack of support from the companies they serve.
The Life of a Data Labeler
Michael Geoffrey Asia’s daily routine illustrates the harsh realities faced by data labelers. For eight hours each day, he sat at his laptop, annotating explicit content for an AI data labeling company. After his official work hours, he transitioned to another job, engaging in sexting as a human operator behind AI sex bots. His work required him to adopt various personas, depending on the demographic of the person he was interacting with.
“It required a lot of creativity and fast thinking,” Asia explained. “If I’m talking to a man, I’m supposed to act like a woman. If I’m talking to a woman, I need to act like a man.” This constant switching took a toll on his mental health, leading to insomnia, PTSD, and difficulties in his personal relationships.
The Impact of Work on Mental Health
After months of this grueling work, Asia reached a breaking point. He described how his body began to fail him, saying, “Where I saw someone naked, I don’t even feel it.” This disconnect not only affected his professional life but also strained his marriage and family life, as he struggled to meet the emotional and physical expectations of his wife.
Asia’s experiences are not unique. Many data labelers report similar mental health challenges, stemming from the nature of their work, which often involves exposure to graphic and disturbing content without adequate psychological support.
The Formation of the Data Labelers Association
In response to these challenges, Asia became the secretary general of the Data Labelers Association (DLA). The organization aims to advocate for better working conditions, including fair pay, mental health services, and the elimination of restrictive non-disclosure agreements. Despite their crucial role in the AI ecosystem, data labelers often earn just a few dollars a day, with little recognition from the companies that benefit from their labor.
Recent Events and Activism
In March 2026, the DLA held a significant event at the Nairobi Arboretum, aimed at recruiting new members and providing a platform for workers to share their stories. The gathering highlighted the need for solidarity among workers in the tech industry, particularly those involved in data labeling and content moderation.
Angela, a speaker at the event, emphasized the historical context of their struggles, drawing parallels between the current labor dynamics and colonial exploitation. “When you think of colonialism, we were under British Imperial East Africa Company,” she stated. “We are just products, part of their operation.” Her call to action urged workers to recognize their position within the global supply chain and to unite in their fight for rights and recognition.
The Global Context of Data Labeling
Kenya has become a significant hub for data labeling, with many tech companies outsourcing this labor to the region. As Asia noted, nearly everyone he spoke to in Nairobi had either worked as a data labeler or knew someone who had. This widespread involvement underscores the critical role that Kenyan workers play in the global AI landscape.
Companies like Sama, a San Francisco-based data annotation firm, have faced scrutiny for their treatment of workers. Despite being contracted by major tech giants like Meta and OpenAI, Sama has been criticized for its low wages and the mental health repercussions faced by its employees. The company’s slogan, “Samasource: THE SOUL OF AI,” starkly contrasts with the reality experienced by its workers.
The Future of Data Labeling and Workers’ Rights
As the demand for AI continues to grow, the need for ethical considerations regarding the labor behind these technologies becomes increasingly urgent. The Data Labelers Association is at the forefront of advocating for the rights of these workers, pushing for better pay and working conditions, as well as mental health resources.
Asia’s journey from a data labeler to an advocate illustrates the potential for collective action among workers in the tech industry. By organizing and sharing their experiences, they hope to bring attention to the often-hidden labor that supports the booming AI sector.
Conclusion
The fight for better working conditions among data labelers in Kenya is a crucial part of the broader conversation about the ethics of AI and the human cost of technological advancement. As these workers continue to organize and advocate for their rights, their stories serve as a reminder of the importance of recognizing and valuing the human labor that powers AI.
Note: The insights shared in this article reflect the experiences of data labelers in Kenya and highlight the need for ongoing discussions about labor rights in the tech industry.
