Is AI Closing the Door on New Graduates?

It has become one of the most repeated anxieties about AI: that the technology is quietly pulling up the ladder behind it, automating away the junior roles where people used to learn their trade. Graduates, the story goes, are being asked to supervise the very systems that were supposed to be their first jobs. It is a genuinely worrying idea, and unlike some AI fears it has real numbers attached. The question is what those numbers actually say.

Start with the strongest signal. A Stanford analysis of ADP payroll data found that employment for 22 to 25 year olds in the most AI-exposed occupations now sits about 19 percent below their less-exposed peers, up from a 13 percent gap a year earlier. Crucially, that gap seems to come mainly through fewer hires rather than layoffs. Firms are not firing their juniors. They are quietly not replacing them. Job-posting data points the same way. Indeed reported entry-level postings down about 7.5 percent year over year while senior postings rose almost 15 percent.

The case for blaming AI, and against it

If that were the whole picture, the fear would be close to confirmed. It is not. A September 2026 working paper from the National Bureau of Economic Research, by Fairlie and Wu, found no statistically significant AI-linked rise in unemployment for recent graduates aged 22 to 25. LinkedIn's own figures muddy things further: entry-level hires are down around 6 percent year over year, but mid-level hires are down about 10 percent. If AI were simply eating the bottom rung, the squeeze would not be landing harder on the middle.

That is the problem with a clean narrative. Hiring has cooled across the board since the pandemic-era surge, interest rates reshaped how companies spend, and remote work changed who gets hired where. AI is a plausible contributor to the entry-level slowdown. It is very hard to show it is the main one, and some of the scariest figures do not survive contact with their sources. A widely shared claim that entry-level hiring fell 80 percent at AI-adopting firms traces back to a working paper and should be treated with real caution rather than quoted as fact.

What is actually true

Here is the honest version. Recent-graduate unemployment, at roughly 5.6 percent, is running above the overall rate of about 4.3 percent, which is unusual and worth taking seriously. A Northeastern survey found that among hiring managers who said AI had changed their decisions, 26 percent were hiring fewer entry-level candidates while only 7 percent were hiring more. The direction of travel is not imaginary.

But the ladder has not been pulled up. It has been made steeper and more crowded at the bottom, by a mix of forces that AI is part of rather than the sole author of. That is a harder message than either the panic or the dismissal, and it is the one the evidence supports. The practical takeaway for graduates is less dramatic than the fear and more useful: the roles that involve directing and checking AI work are growing, even as the old apprentice tasks shrink. We have made the same point about specific professions, from the long-running worry that AI will replace software engineers to the question of whether AI really cut thousands of jobs at Barclays. The pattern keeps repeating. The fear is real, the mechanism is more complicated, and the headline number is usually too tidy to trust.