Artificial intelligence is changing the shape of human work faster than it is eliminating it, according to Brendan Foody, chief executive of the AI hiring platform Mercor, who argues that the technology is creating new categories of jobs even as it automates old ones.
Foody's assessment, delivered in an interview with CBS News, lands as workers across industries press for clarity on which roles will survive the rapid adoption of generative AI tools. Mercor operates as a marketplace that connects companies building AI systems with human workers who perform the tasks those systems cannot yet handle on their own.
The company's business model rests on a counterintuitive premise: the more capable AI becomes, the more it depends on people to train, evaluate, and refine it. That work ranges from data labeling and content review to specialized tasks requiring domain expertise in law, medicine, and engineering.
Foody's argument echoes a pattern labor economists have documented in earlier waves of automation. When ATMs spread across U.S. banks in the 1970s and 1980s, the number of human bank tellers rose rather than fell, because the machines lowered the cost of opening branches. The Bureau of Labor Statistics has similarly projected that demand for workers in AI-adjacent fields, including data annotation and model evaluation, will keep climbing through the decade.
Still, the picture is uneven. A 2023 study by researchers at Princeton University and New York University found that exposure to large language models was highest among workers in telemarketing, legal services, and some clerical roles, jobs where automation could displace workers rather than augment them. Foody's optimism applies most directly to workers whose skills complement AI systems rather than duplicate them.
Mercor has raised venture funding at a valuation in the billions of dollars, reflecting investor bets that human-in-the-loop labor will remain a durable part of the AI supply chain. The company pays contractors for tasks such as ranking model outputs, writing expert reasoning, and correcting errors in generated text.
For workers weighing whether to retrain, Foody's message is that adaptability matters more than picking a single safe occupation. The roles most insulated from displacement, he suggests, are those requiring judgment, verification, and accountability, the tasks companies are least willing to hand to a machine without a human checking the result.
The stakes are substantial. Goldman Sachs economists estimated in 2023 that generative AI could eventually affect roughly 300 million full-time jobs globally, with about two-thirds of U.S. occupations partially exposed to automation. How many of those jobs are transformed rather than terminated will depend on decisions made by employers over the next several years, not on the technology alone.
