UK Data Shows AI May Be Squeezing Computer Science Grads Out of Coding Jobs

The predicted squeeze has a number now. Data published Saturday for the Guardian University Guide 2027 shows the proportion of UK computer science graduates finding professional work as coders or programmers fell from about 40 per cent to just 28 per cent last year. The share moving into any graduate-level occupation dropped too — from more than 60 per cent two years ago to 50 per cent.

The figures come from the Higher Education Statistics Agency’s graduate outcomes survey, which gathered responses from more than 350,000 former students fifteen months after they finished courses completed in 2024. Coding and software development were the fastest-falling occupations for graduates overall, and employer demand for economics graduates in well-paid finance roles — economists, management consultants — declined in parallel.

The researchers’ reading

Matt Hiely-Rayner, director of consultancy Intelligent Metrix, which compiles the guide, was unusually direct for someone working with survey data: “With AI providing such responsive and cheap grunt work in this field, it’s hard not to conclude that this is behind the trend.”

Charlie Ball, head of labour market intelligence at Jisc, the UK’s education data agency, hedged less than most labour-market researchers do. “While I’m reluctant to say that AI did it, because there’s little evidence, it’s very hard to escape the conclusion that AI has played a role,” he said. Both stop short of proof — the survey shows what happened to graduates, not which employers replaced what with which tools. But the mechanism has been predicted for years: AI agents producing software and analysis cheaply remove exactly the entry-level work that new graduates cut their teeth on. We’ve tracked that pipeline argument for months — Yale research on AI hollowing out the entry-level career ladder, and the Stanford/Guardian analysis finding the AI jobs apocalypse hasn’t shown up in aggregate. What’s new here is the data landing on the graduates themselves rather than the job postings.

Not a collapse — a rerouting

Ball’s other observation deserves more attention than it got in most coverage: computer science graduates aren’t vanishing from the workforce, they’re diversifying into adjacent roles like cybersecurity and network engineering. “It’s not collapsed entirely, there are still roles available, but pure coding jobs seem to have been in particularly short supply in 2025.”

The finance side tells a subtler story. Ball talked to employers about their AI use and found most are not replacing jobs outright — they’re changing what the job is, and expecting new hires to arrive already fluent with AI tools. A minority are embracing the tools and laying people off hoping for an edge, which Ball expects may not persist “because everyone is experimenting right now”. That matches what we saw in the iCIMS workforce data on workers self-teaching AI faster than employers train — the skills bar moves before the job count does.

What universities are doing about it

Birmingham’s vice-chancellor Adam Tickell pointed to his university’s response: weaving employer-demanded skills into courses and offering an “intercalated year” in subjects like AI and data science without the usual prerequisites. “This is about equipping people for the world they are in, whether or not we like the world we are becoming,” he said.

For New Zealand the picture is uncomfortably portable. Computer science is among the most popular degree choices at NZ universities, and the local software market skews heavily toward exactly the junior cohort roles — support, integration, maintenance coding — that AI tools absorb first. The UK data won’t map one-to-one, but if the entry-level rung of the programming ladder is thinning out in a market five times our size, the question for NZ students isn’t whether AI affects their job prospects. It’s which adjacent rungs — security, data, networks, AI operations — they should be climbing instead.

❓ FAQ

What do the UK figures actually show? The share of computer science graduates finding professional jobs as coders or programmers fell from about 40 per cent to 28 per cent in the most recent survey year. The broader share entering any graduate-level occupation fell from over 60 per cent to 50 per cent.

Is AI proven to be the cause? No. The researchers themselves hedge — Ball says there’s “little evidence” to say AI did it definitively, but finds it “very hard to escape the conclusion that AI has played a role.” The data shows the trend; the cause is inference.

Does this mean computer science degrees are worthless? The same data shows graduates diversifying into cybersecurity, network engineering, and other roles. Graduate-level employment fell but hasn’t collapsed — half of CS graduates still move into graduate-level occupations.

Why should New Zealand care? NZ’s software industry is heavy on exactly the junior and maintenance coding work this data shows contracting fastest. The UK numbers are an early read on where an AI-absorbing labour market hits the newest workers first.

🔍 THE BOTTOM LINE

For two years the AI-and-jobs debate ran on projections. This is the first UK dataset where the graduates are the datapoints, and the numbers are moving in the direction the AI-will-do-the-grunt-work prediction said they would — with the researchers themselves pointing at AI as the likeliest driver. The honest caveats still apply: correlation, one survey cycle, no employer-level evidence. But “there’s no evidence AI is hurting the job market” just got harder to say with a straight face.