Intelligence

ai

Hidden Underbelly of the Silicon Valley: Algorithmic Exploitation and Health in Data Work Value Chains

Source
arXiv — Computers and Society
Published
Last verified
7 Aug 2026
Confidence
High
Evidence
Original document retained
Reading time
1 min
Country
International
Relevant to
People & Capability, Technology & Data, Research & Evidence

Executive summary

What happened, and why should leadership care?

Research from arXiv highlights concerns regarding the outsourcing of data work by major technology firms, often to the Global South. This practice, critical for AI development, reportedly involves adverse working conditions for data workers who remain largely hidden within value chains. A survey of workers in Kenya, a key hub for such activities, forms the basis of this assessment.

Why this matters

Why is this strategically important?

This issue is strategically important because it exposes the ethical and operational risks associated with current AI development practices and global supply chains. Unaddressed, these concerns can lead to significant reputational damage, regulatory scrutiny, and challenges to maintaining social license to operate for organizations reliant on outsourced data work.

Key insights

What should be noted from the evidence?

  • Large technology firms outsource data work globally, often to regions like the Global South, for AI development.
  • Data workers, particularly in regions such as East Africa (e.g., Kenya), operate under adverse conditions.
  • The extent and nature of these outsourced data work value chains are frequently concealed by firms.
  • This outsourced data workforce is crucial for training advanced AI systems.

Evidence and confidence

How far can this assessment be trusted?

High confidence. Named institution, original document retained and analysis corroborated.

Analysis is prepared editorially by Aziz Shuaib Ausi. The original publication remains the authoritative record, and executive judgement remains entirely human.

Source

Where does this originate?

Reported by arXiv — Computers and Society · International. This briefing summarises the publication for executive use; the document itself is not reproduced here.

Read the original publication