Kenya Sets New Standard with Regulations for AI Workforce

The East African nation is poised to become the first globally to mandate minimum pay and mental healthcare provisions for workers training artificial intelligence systems. This move targets a growing outsourcing industry that has relied on low-cost labor in Kenya.

The proposed policy, outlined by the ICT Ministry, requires companies employing Kenyan AI data workers to adhere to local duty-of-care standards. These include:

  • Safeguards against exposure to harmful content
  • Access to mental health support
  • Transparent contracting practices with clear grievance mechanisms
  • Fair pay benchmarks aligned with international rates for comparable work

Addressing Worker Concerns

The regulations respond to years of complaints from Kenyan workers employed by outsourcing firms serving tech giants like OpenAI and Meta. Content moderators and data annotators often review graphic violence, self-harm, and abuse material while receiving inadequate psychological support.

While some earn slightly above Kenya’s statutory minimum wage (around $1.00 per hour), many receive significantly less than their counterparts in developed countries—where similar roles pay between $21.00 and $27.00 per hour.

The government will establish a fair-pay-reference framework to ensure transparent compensation benchmarks for data annotation, content moderation, and AI quality evaluation roles.

Implications for the Global Tech Industry

Kenya’s emergence as an AI outsourcing hub has been driven by its large English-speaking workforce. Companies have increasingly used specialist contractors in Kenya to reduce labor costs while maintaining legal distance from direct employment relationships.

This regulatory intervention signals a potential shift in how the global AI industry sources its training data, particularly in developing nations where labor protections may be weaker.

The move follows over 35 tech workers filing a landmark legal challenge demanding recognition of exposure to harmful digital content as an occupational hazard requiring specialized care. Kenya’s actions suggest that the era of unregulated outsourcing for AI development is drawing to a close.