The artificial intelligence stock market, which has swollen to over £2 trillion in market capitalisation over the past 18 months, shows clear signs of an imminent correction. This is not the frothy optimism of a healthy sector but the cold calculus of a bubble that has lost touch with fundamentals.
The trigger? A confluence of regulatory headwinds, valuation disconnect, and a startling lack of commercial viability for the most hyped applications. After the stratospheric rise of companies like Nvidia, which saw its market cap triple in a year, the first cracks appeared when the US Securities and Exchange Commission announced a probe into AI chip export controls. Then came the European Union's AI Act, which imposes compliance costs that could slash profit margins by up to 30% for European AI firms.
The City of London has been quietly preparing for this. The Bank of England's Financial Policy Committee warned last month that a sudden reversal in AI equity prices could trigger systemic stress, given the high leverage used by some hedge funds to ride the wave. A senior trader at a major London-based quantitative fund told me: 'We've been reducing exposure to AI since January. The liquidity is drying up and when the sell-off comes, it will be violent.'
The numbers are stark. The Shiller P/E ratio for the AI sector now stands at 45, compared to a historical tech average of 25. Revenue growth has decelerated from 40% to 15% year-on-year for the top ten AI firms, yet share prices still command a 20% premium over the broader market. This is not sustainable. The real danger is not a gradual decline but a flash crash where algorithmic trading systems, now accounting for 70% of daily volume, trigger a cascade of stop-loss orders.
What does this mean for the average person? Pension funds heavily invested in AI index funds could see a 30% haircut, wiping out years of returns. The UK government, which has championed AI as a cornerstone of its economic strategy, may find its tax receipts and growth forecasts upended. And for the startups built on the promise of AI, venture capital will dry up, leading to a wave of layoffs and closures.
But the most unsettling aspect is the user experience of society. If the bubble bursts, the narrative that AI will solve all our problems will implode. We will be left with the real, less glamorous work: making AI safe, equitable, and genuinely useful. The prophets of the singularity will be silenced, and in their place, we will need engineers, ethicists, and regulators who understand that technology must serve humanity, not the other way around.
As I write this, the VIX index has spiked 40% in a week, bond yields are rising, and gold is hitting new highs. The fear is palpable. The question is not if the bubble will burst but when and how badly. For the City of London, the reckoning is near. And for those who have bet the house on AI magic, the hangover will be brutal.
