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

• A version of this story originally appeared in the BI Tech Memo newsletter.
• Sign up for the weekly BI Tech Memo newsletter here.

As AI systems grow more capable, the bar for human training has risen sharply, and generalist data labelers are being pushed aside.

That's according to HireArt's 2025 AI Trainer Compensation Report, which collected information from more than 150 sources, including a survey of active workers, public job postings, and internal data.

The study shows that today's AI models demand nuanced reasoning, domain expertise, and multilingual fluency, transforming "data labeling" into specialized cognitive work.

Subject-matter experts in fields such as law, engineering, and medicine now command dramatically higher rates than traditional data annotators, reflecting their new strategic importance.

The shift signals a maturing industry where high-skill trainers, not crowdsourced gig workers, increasingly shape model intelligence, safety, and trustworthiness.

Entry-level, generalist AI model trainers earn $12.50 to $15.50 in the US, while highly skilled expert AI trainers can make more than $100 an hour, depending on the subject, according to the data compiled by HireArt.

In the medical field, top experts can make $60 to over $180 an hour. In engineering and law, hourly pay ranges from $80 to more than $150, HireArt data shows.

The chart below shows the midpoint of these pay ranges. You can see how much more highly-skilled experts can earn. As it should be!

Sign up for BI's Tech Memo newsletter here. Reach out to me via email at abarr@businessinsider.com.