by This week, America’s tech powerhouses are under the spotlight as they report their July–September results. Microsoft, Alphabet (Google’s parent), Amazon and Meta Platforms have all posted brisk revenue growth for the quarter, according to data from LSEG.
Yet, amid the strong numbers, one big concern is looming: has the artificial-intelligence (AI) boom that has propelled valuations in recent years run ahead of fundamentals — and is a bubble forming?
The companies themselves are expected to emphasise their ongoing investments in AI, signalling their belief in its long-term promise. But industry leaders from within and beyond these firms are increasingly warning that the frenzy may be outpacing real performance.
Signs of Strength — and Sources of Skepticism
What’s encouraging
-
The cloud units of Microsoft, Amazon and Google are expected to show strong growth in the quarter. For example, Microsoft’s Azure unit is forecast to grow by ~38.4%, Google Cloud ~30.1%, and Amazon Web Services ~18%.
-
Overall company revenue growth for the quarter is also robust: Microsoft ~14.9%, Alphabet ~13.2%, Amazon ~11.9%, Meta ~21.7%.
-
These firms maintain huge balance sheets and appear willing to keep funding AI infrastructure, which gives them the capacity to play the long game.
Why there are warning flags
-
A widely cited study from the Massachusetts Institute of Technology (MIT) found that of more than 300 AI projects analyzed, only about 5% delivered measurable gains. Most were stalled or failed to scale.
-
Some analysts point to unusual deal structures and financing models. For example, tech firms investing in hardware, infrastructure and each other in circular ways — reminiscent of the dot-com era.
-
Valuations of many AI-related firms have detached from their current earnings or proven profit models, but instead rely heavily on future expectations of what AI might deliver.
Thus, while the AI investment story remains bullish, there is increasing caution: are the fundamentals keeping pace with investor optimism?
What’s Driving the Bubble Talk?
Massive spending, unclear returns
Tech giants and cloud firms are slated to spend around US$400 billion this year on AI infrastructure alone.
But the major question: when will these investments translate into reliable, scalable business returns? The MIT study suggests most early projects fail to show the hoped-for payoff.
Circular deals and growing debt
There’s been a notable increase in complex and inter-linked deals among AI ecosystem players. For example, firms who are customers of others also invest in them, build infrastructure for them, or depend on them.
And rather than being purely funded internally, some of the AI build-outs are being financed by new borrowing — shifting the risk profile.
Valuations built on future potential, not current output
Many valuations in the AI space hinge on very optimistic future outcomes, rather than proven revenue streams today. One academic paper introduced the “Capability Realisation Rate” (CRR) model to quantify this gap between expectation and actual performance.
As one writer put it: “Valuations for AI companies are based on aggressive projections rather than current earnings.”
Investors are uneasy
Some prominent investors and firms are explicitly warning of risk. For example:
-
The Bank of England’s Financial Policy Committee noted rising risk of a sharp market correction, especially if AI expectations aren’t met.
-
Investor consensus is leaning toward “we may not be in the bubble crash stage yet, but conditions are getting closer.”
Why Some Still Believe the AI Boom Has Legs
Even amid caution, there are compelling arguments in favour of the AI investment wave continuing:
-
Some investors point out that adoption may be relatively low now, but the potential for growth is large. In other words: current results may lag but future payoff could be strong.
-
Big-tech companies involved in AI still report strong cash flow, and they’re backing their bets — which gives credence to longer-term visions.
-
Compared with past tech booms, this wave is somewhat more concentrated: it’s largely around AI and a few major firms, rather than a broad retail tech boom. Some analysts argue that means a potential correction may be smaller or more limited in scope.
So What Could Happen?
Scenario 1: A correction or pull-back
If AI projects continue to struggle with scaling and integration, investor sentiment might shift. Over-valued companies might see sharp de-ratings. Circular financing and debt may amplify the risks. In such a scenario, the tech sector’s heavy reliance on AI-driven growth could leave it vulnerable.
Analysts warn the next 12-24 months could bring a “drawdown” in the equity markets tied to AI hype.
Scenario 2: A slower-burn rather than sudden crash
Given that much of the investment is concentrated in a few large firms with strong balance sheets, any adjustment might be more gradual than the dot-com bust. The infrastructure build-out may take years to pay off, and value may emerge more slowly. Some argue the story’s still positive, just delayed.
Scenario 3: Surge into actual value
If key business models finally scale, integration becomes smoother and AI becomes embedded across the economy, many of the current investments could be vindicated. The key will be demonstrated returns on large-scale AI adoption rather than just pilots or hype.
What It Means for Companies & Investors
For companies reporting results: the spotlight is very much on execution. Generative AI tools, cloud-infrastructure build-out, and capacity to monetise AI will be the differentiators. Revenue growth is good, but cost pressures and margin impacts will also matter. For example: even as revenue growth is strong, profit growth for some of the major companies is expected to soften due to higher costs.
For investors: the key question is how much of the current valuation is based on near-term reality versus future hope. Questions to ask:
-
Are expected business models delivering?
-
Are projects moving beyond pilot to scale?
-
Is the capital spend being matched by revenue growth or real productivity gains?
-
What happens if AI costs, infrastructure needs or regulatory hurdles increase?
Final Thoughts
The current AI surge is one of the defining narratives of 2024–2025 in tech. It has powered revenue growth, excited investors and drawn large capital commitments. But the foundations beneath it are not yet uniformly proven. The line between a boom and a bubble depends on how well the promise of AI turns into measurable business value and sustainable earnings.
The large-cap tech players remain in a strong position: they have resources, brand, infrastructure and scale. But even for them, the margin for error is shrinking. As they report earnings and lay out their capital plans, how many of these AI bets pay off — and when — will determine whether we view this era as a transformative shift or a cautionary episode in tech investing history.
