The AI skills gap is here, says AI company, and power users are pulling ahead

Anthropic’s latest research suggests that while AI is rapidly changing the way work is done, it has not eliminated meaningful jobs. At least, not yet. But beneath what Anthropic’s head of economics, Peter McCrory, says lies a “still healthy” labor market, early signs point to uneven impacts, especially for younger workers just entering the workforce.

In an interview on the sidelines of the Axios AI Summit in Washington DC, McCrory said the company’s latest economic impact report so far shows little evidence of widespread job displacement.

“There is no material difference in unemployment rates” between workers who use Claude for the “most central task of their jobs in an automated manner” – such as technical writers, data entry clerks and software engineers – and workers in jobs that are less exposed to AI and require “physical interaction and dexterity with the real world.”

But as AI adoption spreads across industries, that could change quickly. If Anthropic CEO Dario Amodei is to be believed, AI could do just that wipe half away of all entry-level white-collar jobs, which could see unemployment reach 20% in the next five years.

“Displacement effects can manifest very quickly, so you want to put a monitoring framework in place to understand this before it manifests so we can catch it as it happens and ideally identify the right policy response,” McCrory told TechCrunch.

Following these trends is why tracking the growth, adoption and spread of AI is so important, he said.

In theory, McCrory says, AI models like Claude can do almost anything a computer can do. In practice, most users are just scratching the surface of these possibilities.

He said Anthropic looked at which roles include tasks that AI is particularly good at, that are already being automated, and that relate to real-world workplace use cases – the areas most likely to indicate where displacement could occur.

Anthropic’s fifth economic impact report, released on Tuesday, also found that even where there has not yet been much displacement, there is a growing skills gap between previous Claude users and newcomers.

It is more likely that previous users will get significantly more value from the model, using it for work-related tasks rather than informal or one-off purposes and in more sophisticated ways, such as as a ‘thought partner’ for iteration and feedback.

McCrory said the findings suggest that AI is becoming a technology that rewards those who already know how to use it – and that workers who can integrate it effectively into their work will increasingly have an advantage.

That advantage is also not evenly distributed geographically. The report also shows that “Claude is used more intensively in high-income countries, in the US in places with more knowledge workers, and for a relatively small number of specialized tasks and occupations.”

In other words, despite AI’s promises as an equalizer, adoption may already be tilted toward the wealthy and these benefits could amplify as the users of power advance.