Copy article

Are we facing a Dotcom-style AI crash? "Foolish to think this time is different"

ended 03. December 2025

FINANCIAL experts have warned it is “dangerous to think this time is different” following the Bank of England warning of a possible “sharp correction” in the value of AI companies and the global economy's exposure to a multi-trillion dollar debt mountain.

In its Financial Stability Report, the Bank of England said: “Markets do not appear to be fully reflecting the persistent material uncertainty in the global economic and policy environment, and the potential for adverse outcomes. Some risks are difficult to price, either because they are very unlikely to occur or their impacts are highly unpredictable. The risk of a sharp correction remains high.

"If participants were to reassess their outlook abruptly this could cause asset prices to realign to the prevailing high level of uncertainty.

"The risk of a sharp correction sits against a backdrop of rising public debt-to-GDP ratios in many advanced economies, potentially constraining their governments’ capacity to respond to future shocks.

“Notably, equity market valuations for technology companies focused on artificial intelligence (AI) remain materially stretched. On some measures, equity valuations are close to levels not seen since the dot-com bubble in the US, and the global financial crisis in the UK.”

  given the trillions of dollars of debt and investment invested in AI stocks,  role of debt financing in this [the AI} sector is increasing quickly as AI-focused firms seek large scale infrastructure investment. Deeper links between AI firms and credit markets, and increasing interconnections between those firms, mean that, should an asset price correction occur, losses on lending could increase financial stability risks.

 

The Bank of England has finally said the quiet part out loud: the AI market looks dangerously like a bubble ready to burst. In its latest Financial Stability Report, the Bank explicitly warns of a "sharp correction" with valuations looking "stretched" in ways that bring back nasty memories of the dotcom crash.

We are looking at a sector projecting $5tn in infrastructure spending, much of it fuelled by debt, while many businesses are still struggling to find use cases that deliver actual profit rather than just promise. Governor Andrew Bailey notes that unlike the dotcom era, these firms have cash flow, but he acknowledges the sector is frighteningly concentrated.

We have seen this movie before. Throwing trillions at a mostly unproven 'black box' churning out slop doesn't create value; solving human problems reliably does. 

When half of that infrastructure spending is funded by external debt, we aren't just looking at a tech correction; we are looking at a financial stability risk that could hit the wider UK economy hard.

We’d like your views:

  • Is the projected $5tn infrastructure spend a genuine investment in productivity or just the world's most expensive FOMO exercise?
  • The BoE draws direct parallels to the dotcom bubble. Are we ignoring the lessons of history in favour of hype?
  • If the bubble bursts, will it finally force a return to practical, human-centred automation rather than generative AI fantasy?
  • With massive debt fueling this sector, how exposed are ordinary UK businesses to this "sharp correction"?
  • Are we seeing a productivity revolution, or just a concentration of wealth in a few US tech giants?
  • Like banks, these mega tech corps are too big to fail. What does that mean if and when they do?

6 responses from the Newspage community

Copy all

Star Quote
Copy

Editor
Antonia Medlicott
Founder & MD at Investing Insiders
It would be foolish to think this time is different to the Dotcom collapse, when history tells we are most likely on course for a correction. It's tempting to think otherwise when you're riding a high of huge price growth and spectacular returns, though. There will be people trying to time their trades, seeking to sell at the top, and buy again at the bottom, to benefit from any potential bounce back. But it's important to note perfect timing is rare and those who achieve it are often just lucky. Investors should be less concerned with "timing" this market, and more focused on gaining "time in the market". Staying invested for the long-term allows you time to benefit from a recovery, which history suggests will always follow with enough time. A diversified portfolio, not solely reliant on AI and not concentrated too heavily on US stocks, should also help to smooth out the effects of any market shock in one particular sector or geographic region.
Copy

Editor
Rob Mansfield
Independent Financial Advisor at Rootes Wealth Management
It's different this time is a dangerous adage but drawing too many parallels with the Dotcom crash may also be dangerous. At that point some companies were being valued by how many webpage visits as they weren't profitable. Now that's true now of some companies but the likes of Nvidia, Meta, Alphabet are incredibly profitable and so may be better placed to sustain these risks. That being said, valuations do look stretched and these valuations are relying heavily on AI being the next best thing since sliced bread. It wont take a lot to cause a wobble and then we'll see how robust this technology is.
Copy

Editor
Tony Redondo
Founder at Cosmos Currency Exchange
Part of the Bank of England's role is risk control, and viewing the AI sector through a historical lens raises concerns about whether AI companies are becoming too big to fail, hence Threadneedle Street's stark warning. There are uncomfortable parallels with the dotcom bubble, though one key difference stands out: AI companies already generate substantial revenue, whereas many dotcom companies had zero revenue and were valued purely on "eyeballs" and traffic. The real concern for me isn't economy-wide productivity gains, but rather the wealth concentration in a few US giants, which is the ultimate K-shaped economy.
Copy

Editor
Anita Wright
Chartered Financial Planner at Ribble Wealth Management
The BoE’s warning about an AI-driven “sharp correction” sits on top of a far larger structural problem: we are not just speculating on a new technology, we are layering a vast, debt-financed bet on AI onto a monetary and fiscal system that is already at breaking point. Globally, the major central banks – led by the Federal Reserve – are effectively in a regime of fiscal dominance. Western social entitlement systems have taken on the character of a Ponzi structure: more is being taken out than paid in, and the mathematics no longer work without constant financial repression. Now into this mix comes AI. The headline number – around $5 trillion of infrastructure spending, a large share of it funded by external debt – is being sold as a productivity revolution. Yet the macro frame is awkward. AI is extraordinarily capital-intensive, demands huge upgrades to power grids and data infrastructure, and arrives at a moment when debt-to-GDP ratios are already at, or above, post-war extremes.
Copy

Rohit Parmar-Mistry
Founder at Pattrn Data
$5tn on infrastructure is the world’s most expensive case of corporate FOMO. The Bank of England has finally said the quiet part out loud: we aren’t building a productivity revolution; we are building a massive data centre for solutions nobody asked for. The Governor’s parallel to the dotcom bubble is spot on, but the reality is more insidious. We are funding a global gambling addiction with debt the real economy cannot afford, all concentrated in a handful of US giants. I see this daily on the ground: businesses drowning in 'black box' tools that create work rather than reducing it. They are betting on generative fantasy rather than practical utility. This $5tn spend isn't innovation; it is a casino selling solutions to problems that do not exist. If this bubble bursts and the "stretched" valuations suggest it will, it might finally flush out the hype merchants. We need to stop treating AI like a lottery ticket and return to boring, practical automation that actually helps humans.
Copy

Colin Low
Managing Director at Kingsfleet
It has been said that the four most dangerous words in economics are “This time it’s different”. For the last few years, investment experts have highlighted the extended valuations of megacap US tech businesses and yet, month after month, their capitalisation has increased, without any sense of rationality. At the root of this is the growth in index investing because there is no differentiation in a stock representing good or poor value. Whilst index investing keeps costs down, investors must understand that their basket may consist of stocks that can be highly overvalued.