As AI technology evolves, the need for highly specialized human trainers has surged, transforming how models like ChatGPT and Cohere are trained. In earlier stages, training relied heavily on large teams of low-cost workers to help AI distinguish between basic facts, such as identifying images of cars versus carrots. However, with advancements in AI models, companies now seek experts in fields like history, medicine, and finance, often employing individuals with advanced degrees.
Cohere co-founder Ivan Zhang notes, “A year ago, we could hire undergraduates to teach AI. Now, we have licensed professionals in specialized fields guiding the models.” Companies like Invisible Tech, which partners with multiple AI firms, have emerged as crucial players in this landscape, employing around 5,000 trainers across over 100 countries.
Invisible Tech’s founder, Francis Pedraza, highlights their expertise: “We work with PhDs and specialists, paying rates that can reach $200 per hour for highly complex subjects.” Initially founded as a workflow automation company, Invisible pivoted to AI training after collaborating with OpenAI, addressing issues like AI hallucinations—instances where the AI generates incorrect or misleading information.
To combat hallucinations, AI companies are increasingly enlisting human trainers to impart the nuances of truth and fiction. OpenAI, a pioneer in generative AI, employs a dedicated team known as the “Human Data Team” to work alongside these trainers, refining the model’s abilities and improving output quality.
As competition grows in the AI sector, the demand for skilled trainers has led to a rise in firms offering similar services. Zhang notes, “My inbox is flooded with new companies emerging in this space.” This trend indicates a burgeoning niche market where professionals can contribute to AI development without needing coding skills, driven by the escalating complexity of AI applications.
