AI BUBBLE?
AI’s trillion-dollar question just got louder
Three years after ChatGPT sparked the AI boom, the money is still flowing, but investor confidence is wobbling.
Drops in Nvidia and Oracle shares, plus concerns about companies linked to OpenAI, have raised questions about whether AI’s rapid growth can keep up with its rising costs.
The core issue is scale: AI is extremely expensive to build, and it’s not yet clear whether long-term demand will justify the investment.
Because big tech firms have powered much of the stock market’s recent gains, any slowdown from them could ripple out.
OpenAI alone plans to spend $1.4 trillion, despite earning far less than it spends and expecting losses until 2030.
Funding hasn’t been a problem so far, but a shift in investor appetite could affect the wider AI ecosystem.
Oracle’s recent decline shows how sensitive the market is to higher capital spending and delays in AI infrastructure.
At the same time, Alphabet, Microsoft, Amazon and Meta are set to spend more than $400 billion on data centres next year.
Revenues tied to AI are rising, but not at the same pace. Depreciation costs are accelerating, and some companies may struggle to maintain buybacks and dividends.
Here’s what you should know:
AI investment is huge, but confidence is starting to cool.
Big Tech’s spending is rising faster than related revenue.
Valuations are high but not at bubble levels.
The trillion-dollar tightrope
Valuations remain elevated but still well below dot-com-era extremes.
The Nasdaq 100 trades around 26 times earnings, compared with more than 80 during the bubble.
Some firms have stretched multiples, but major players like Nvidia, Microsoft and Alphabet remain comparatively restrained.
For now, investors face a mixed picture: high costs, slower growth forecasts, and strong long-term belief in the technology, but no clear signal about where the AI trade heads next.
I’ve got a data centre for sale if anyone wants it? - MV