Middle England has been told, day after day, that the great threat to its living standards is illegal migration. Pressure on wages. Pressure on schools. Pressure on the NHS. Pressure on housing. A sense that the ladder of opportunity is being pulled up for their kids.
It’s a compelling story because it comes with a ready-made villain. You can picture it: a small boat, a hotel bill, a queue at A&E. It is visible, immediate and politically potent.
But while the country argues about what it can see, something far bigger has walked straight through the front door — and it didn’t need a passport. Quite simply, it’s artificial intelligence.
Not the Hollywood version. The dull, efficient, spreadsheet-and-email version. The “write me a report”, “summarise these documents”, “draft the contract”, “screen these CVs” version. The kind of work that used to be the first rung on the ladder for bright graduates prepared to graft and prove themselves.
Here’s the uncomfortable truth. For the average British family, for the junior professional, for parents who have paid for degrees and now watch their children compete for scarce internships, immigration is not the main force squeezing pay and prospects. AI is.
Decades of research show immigration’s overall impact on wages and employment is generally small. Where effects do appear, they tend to be localised and sector-specific rather than a blanket wipe-out of middle-class security. Even the fiscal impact is typically measured at well under 1% of GDP. Net migration has also fallen sharply in the latest figures.
None of that means illegal migration isn’t real or politically explosive. It is. But if you are worried about your next promotion, or whether your son or daughter will secure a decent graduate job, immigration is not the silent hand removing rungs from the ladder. AI is.
AI doesn’t sleep. It doesn’t ask for sick pay. It doesn’t join a union. It doesn’t need a desk. It scales at the click of a button. Crucially, it is very good at the exact tasks that once justified hiring junior staff.
Graduate hiring has fallen significantly in recent years. Major employers have trimmed recruitment, and entry-level listings have dropped to levels not seen in a decade. This is not a blip. It is structural change.
What’s disappearing are the foundations of early careers: first drafts, basic research, initial analysis, routine coding, document reviews, slide decks, compliance checks — the endless administrative work that once allowed young people to learn on the job. White-collar, degree-heavy roles are among the most exposed, and AI adoption is strongest in the service sectors where so many middle-class careers sit.
This is where the old political playbook falls apart. You can tighten visa rules. You can patrol borders. You can deport people. But you cannot deport an algorithm.
Immigration pressures, when they exist, tend to show up in specific places — housing demand in overheated markets, stretched local services, competition in certain job segments. Those issues deserve serious, grown-up policy responses.
AI is different. It is not local; it is systemic. If a firm can replace ten junior analysts with one experienced manager supported by AI tools, the entire career ladder shifts. There are fewer entry points, slower progression and tougher competition for what remains. Wages at the bottom of the professional pyramid come under pressure.
That is how the middle class gets hollowed out — not through one dramatic event, but through the quiet erosion of the starter jobs that once turned education into earnings, and earnings into security.
If you want a more accurate image than a boat in the Channel, picture an inbox being cleared in seconds by software. That is where the real disruption is taking place.
This is not an argument for smashing the machines or pretending we can freeze technology in 1995. AI will boost productivity. It will create new businesses. It will make some firms leaner and more competitive.
The question is whether we manage the transition properly or simply let it rip and hope for the best.
We need a serious plan for entry-level work, including incentives for businesses to maintain genuine junior roles, apprenticeships and paid training rather than using AI purely as a blunt cost-cutting tool. Schools and universities must double down on what machines struggle with: judgement, accountability, client relationships and real-world problem-solving. Employers, too, need to rediscover the lost art of training people rather than expecting ready-made talent.
There should also be honesty. If large organisations are making productivity gains through AI, they should be transparent about what that means for early-career opportunities. And if AI boosts profits while shrinking payrolls, it is reasonable to debate how some of that upside can support reskilling and strengthen the broader economy.
Middle England is right to worry. Living standards matter. Opportunity matters. But it is time to point the finger in the right direction. The border debate will continue because it is visible and politically convenient. Meanwhile, the real transformation of work is already under way.
Handled badly, AI could narrow opportunity and entrench inequality. Handled well, it could make British businesses more competitive and create better, higher-value roles for the next generation.
The choice is not between open borders and closed borders. It is between drifting through a technological revolution or leading it. And that is a fight worth having.
