
TL;DR
AI companies are recruiting highly educated professionals to teach machines the judgment they developed over years, even when those systems may eventually replace them.
James Maisiri has a Ph.D. in industrial sociology from the University of Johannesburg. After graduating this year, he expected to teach at a university. Instead, his first job offer came from a recruiter, inviting him to train an AI system to design assessments, teach undergraduates, and grade essays. In short, to hand over a decade of learning to a machine.
He went to the interview. Why would anyone do this? Because South Africa's reality is brutal: the pay was 600 rand ($37) an hour, 20 times the legal minimum wage, while youth unemployment stood at 47.4%. When making a living collides with your principles, which wins?
He is not the only one facing that choice
Around the world, 'human teachers' are being hired to train the AI that will replace them: writers are paid low wages to 'humanize' AI-generated text; in India, waste sorters and welders strap cameras to their heads to record their daily motions for $2.60 an hour, data used to train humanoid robots for their own jobs.
These people occupy different rungs on the pay scale, but the job is the same: transferring human knowledge and judgment to machines. The author reflects that it took him thousands of hours to learn to grade essays, to tell memorization from understanding, to decide why one paper merits 75% and another 60%. The AI would learn not just what he knew, but how he decided.
An AI interviewer decides who gets hired
Professions defined by discretion — lawyers, doctors, engineers — are now being aggressively recruited by AI platforms like Outlier, Mercor, and Surge. The author's own hiring process involved no human: an AI interviewed him for 45 minutes, emailed him feedback on his strengths and weaknesses, and suggested he retake part of the assessment. He declined, and walked away from the job.
He still cannot fully explain why. Perhaps it was the fear of teaching a machine to replace the work he does. He continues to look for an academic post, teaching students, not AI.
Curated from high-quality sources, with concise summaries and key takeaways.