The City is abuzz with talk of artificial intelligence, and it carries the unmistakable whiff of a bubble. Investors have piled into AI-related equities with the enthusiasm of a novice punter at Ascot, driving valuations to levels that would make even the most bullish analyst blush. Nvidia, the poster child of this revolution, has seen its market capitalisation balloon past $2 trillion, a figure that defies historical precedent and common sense.
But as any seasoned trader will tell you, what goes up must come down. The current mania is eerily reminiscent of the dot-com bubble, when companies with little more than a '.com' in their name were valued in the billions. Today, it's any firm that whispers 'AI' into its earnings call. The question is not whether we are in a bubble, but when the pin will prick it.
Let's talk fundamentals. Nvidia's price-to-earnings ratio hovers near 70, a level that implies decades of hypergrowth already priced in. Meanwhile, its revenue, though impressive, is tied to hardware sales that are inherently cyclical. The AI industry is still in its infancy, and the path to profitability for many of its players is littered with technical hurdles, regulatory risks and a looming energy crisis. The data centres that power these models consume electricity like a Victorian factory, and governments are beginning to eye carbon taxes.
Then there's the spectre of central bank policy. The Bank of England and the Federal Reserve have been hiking rates to tame inflation, and the cost of capital is no longer free. The era of 'money for nothing' is over. When the cost of borrowing rises, speculative assets tend to be the first to suffer. We are already seeing signs of capital flight from high-growth tech into defensive sectors. A rotation is underway, and it is not friendly to AI stocks.
Gilt yields are another canary in the coal mine. The 10-year yield has climbed above 4% in the UK, reflecting expectations of sustained inflation and tighter monetary policy. Higher yields make future earnings less attractive, and they suck liquidity out of equity markets. If this trend continues, the AI sector will face a brutal re-rating.
Governments, too, are waking up to the risks. The EU's AI Act and the UK's upcoming regulatory framework could impose compliance costs that squeeze margins. Meanwhile, the US is mulling export controls on advanced chips, which could disrupt supply chains. This is not a sector that operates in a regulatory vacuum.
Of course, the true believers will argue that AI is a transformative technology on par with the internet or electricity. Perhaps it is. But even the most transformative technologies have their boom-and-bust cycles. The railway mania of the 1840s brought us a network that reshaped the world, but it also wiped out countless investors. The same will happen here.
The bottom line is this: markets are forward-looking, and they have already priced in a utopian future for AI. The slightest disappointment, be it a missed earnings target or a new regulation, could trigger a sell-off of epic proportions. The trillion-dollar boom looks increasingly like a casino, and the house always wins.
My advice? Tread carefully. The bulls will tell you this time is different. It never is.
