AI model training needs are changing, with subject experts pushing aside generalist data labelers

As artificial intelligence systems advance, the requirements for human trainers have evolved significantly. A recent report from HireArt highlights that generalist data labelers are increasingly being replaced by subject-matter experts. This shift reflects a growing need for nuanced reasoning, domain-specific knowledge, and multilingual capabilities in AI model training. The findings, derived from over 150 sources, including surveys and public job postings, illustrate that the role of data labeling has transformed into a specialized cognitive task. Experts in fields such as law, engineering, and medicine are now commanding considerably higher rates compared to traditional data annotators, underscoring their critical role in enhancing model intelligence and safety. According to the report, entry-level AI trainers in the United States earn between $12.50 and $15.50 per hour. In stark contrast, highly skilled trainers can earn upwards of $100 per hour, depending on their area of expertise. For instance, in the medical sector, top professionals can earn anywhere from $60 to over $180 per hour, while those in engineering and law can receive between $80 and $150 per hour. This substantial pay gap highlights the increasing value placed on specialized knowledge in the AI training landscape.