The $10 Billion Startup Training AI to Replace the White-Collar Workforce

Editor 19 Apr, 2026 ... min lectura

As artificial intelligence rapidly integrates into the global economy, a new wave of innovation is reshaping the future of work. At the forefront of this transformation is a company called Mercor, a startup valued at over $10 billion that is redefining how human expertise is leveraged to train the next generation of AI systems. Unlike traditional tech companies that focus on developing algorithms and models, Mercor operates on a unique model: it recruits professionals from diverse fields to become 'AI trainers'—individuals who teach AI systems to perform complex tasks that require human-like judgment and contextual understanding.

Founded in 2022 by a group of young entrepreneurs influenced by Peter Thiel’s vision, Mercor has quickly become a benchmark for AI training platforms. Its approach, which focuses on 'human-in-the-loop' augmentation, has attracted attention from governments, corporations, and academic institutions worldwide. But beneath the glossy surface of this growth lies a critical paradox: as Mercor trains AI to replace white-collar jobs, it simultaneously creates new vulnerabilities for the very professionals it hires.

How does Mercor work? A deep dive into the training process

Mercor’s training process is designed to simulate real-world scenarios where AI must adapt to unpredictable human behaviors and ethical dilemmas. Trainers at Mercor act as 'human supervisors' who guide AI systems through complex decision-making processes, such as interpreting nuanced language, resolving conflicts, and applying ethical frameworks in high-stakes environments. This hands-on approach allows AI to learn from human expertise, which is critical in fields like legal, medical, and financial decision-making.

What makes Mercor different from other AI training platforms?

  • They use a 'human-in-the-loop' model that emphasizes contextual understanding over simple pattern recognition.
  • Trainers are paid hands-on to ensure that AI systems can handle real-world ambiguity.
  • Mercur’s training data is curated through a network of professionals who have expertise in high-stakes decision-making.

The company’s rapid growth has raised questions about the sustainability of this model. With over 1.2 million trainers now active globally, Mercor’s impact on the labor market is profound. Yet, as the AI systems it trains become more sophisticated, the risk of displacement for these very trainers grows. This is the central tension of Mercor’s story: a company that creates AI to replace white-collar jobs, while simultaneously creating new roles for those same professionals.

Experts warn that without proper safeguards, this model could lead to a 'double displacement' effect. Trainers who are trained to teach AI systems may find themselves displaced as AI becomes more autonomous. This is not a hypothetical concern—it’s already happening in the early stages of Mercor’s expansion.