In a recent exposé, The Verge has shed light on the often overlooked human labor that powers artificial intelligence (AI) systems. The report focuses on the role of data annotators, the individuals who process and label the raw data used to train AI models.
The story centers around a 30-year-old man from Nairobi, known as Joe, who began his career as an annotator, labeling footage for self-driving cars. His work involved identifying every vehicle, pedestrian, cyclist, and other elements a driver needs to be aware of, frame by frame, from every possible camera angle. This laborious and repetitive task earned him about $10 for eight hours of work.
Joe’s career took a turn in 2019 when he was offered the opportunity to run an annotation boot camp for a new company called Remotasks. This company is a subsidiary of Scale AI, a multibillion-dollar Silicon Valley data vendor. However, most of the annotators, including Joe, were unaware of this connection until recently.
The report highlights the tedious, repetitive, and often underpaid work of annotators, who are a crucial part of the AI development process. It also discusses the secrecy surrounding the work, as annotators often don’t know what they are training AI to do.
The Verge’s investigation concludes by discussing the future of AI and the potential for AI to change the nature of work. It suggests that AI may not replace jobs but change how work is organized, leading to a future where work might become more alien, more isolating, and more tedious.