The Skills Gap That Isn't: London's AI Excuse

That's according to a BusinessLDN survey, which the BBC dutifully reports as evidence of a genuine shortage. Let us, for a moment, entertain the alternative explanation: that this 'skills gap' is a self-serving fiction, a convenient fig leaf for companies that have slashed training budgets and now expect employees to magically acquire capabilities they were never paid to learn.

Now the same employers turn around and say they can't find workers with AI skills. The symmetry is almost too perfect: first you fire people, then you claim nobody qualified exists. It is a neat rhetorical loop, and it requires a credulous press to sustain it. The BusinessLDN survey, for all its earnest concern, does not ask the obvious question: how many of these firms have actually invested in retraining their existing staff? The answer, I suspect, is vanishingly few.

Consider the counter-evidence from the other side of the same news cycle. Forbes published a piece this week arguing that AI layoffs 'don't mean what you think' — that companies are merely reallocating resources, not replacing workers outright. Even if we take that sunny interpretation at face value (and I do not), it still means thousands of people are being shoved into job markets where employers claim the skills aren't there. The logical conclusion is that the 'gap' is a demand-side problem dressed up as a supply-side failure. Companies want instant expertise without the bother of developing it, and when they can't find it, they complain rather than build.

The BusinessLDN report calls for increased investment in training and education. How daring. This is the perennial cry of industry bodies that spend the other 364 days of the year lobbying against exactly that — because training costs money, takes time, and might, heaven forbid, require executives to admit they don't know what they're doing either. The AI boom has turned competence into a commodity: you either have it or you don't, and if you don't, you're fired. No company wants to be left holding the bill for upskilling a workforce that might leave for a competitor next quarter. So instead, they externalise the problem, blaming schools, universities, and government for failing to produce a pipeline of AI-ready workers.

But let's be honest about what this 'AI-ready worker' looks like. The skills in question are not all cutting-edge machine learning research. Most London businesses need people who can use generative AI tools, interpret data, and integrate automation into existing workflows — things that can be taught in weeks, not years. The real barrier is not cognitive but structural: companies refuse to treat training as an investment. They would rather fire ten people and hire one 'AI specialist' at triple the salary, then cry skills gap when that specialist leaves within a year. The revolving door is a feature, not a bug, of an industry that values short-term metrics over human capital.

To be clear, I am not denying that genuine AI expertise is scarce. It is. If half of London's businesses truly lack the skills they need, they have only themselves to blame. The skills gap is a mirror: look into it and you'll see a boardroom full of executives who preferred layoffs to learning.