What is an AI Bubble and How Will it Impact Enterprise AI?

The AI industry is entering what a growing number of analysts are calling ‘bubble’ territory.

With investment surging to levels that echo the heights of the late 1990s dotcom boom, UBS, the global investment bank and financial services provider, is compiling research.

Its findings highlight both the risks and wider ramifications of this emerging ‘AI bubble’ as it surfaces in more and more business discussions across the world.

What is an AI bubble?

An AI bubble describes a market environment where companies building AI technologies are valued significantly higher than their existing revenues and established business models justify.

Such conditions emerge when investor excitement about AI’s transformative potential pushes stock prices and funding decisions to be driven by expectation rather than proven performance.

The vast sums currently flowing into the sector place these concerns into sharper perspective.

Sam Altman, CEO of OpenAI sums it up when speaking to Business Insider: “When bubbles happen, smart people get overexcited about a kernel of truth. Are we in a phase where investors as a whole are overexcited about AI? My opinion is yes. 

“Is AI the most important thing to happen in a very long time? My opinion is also yes.”

The bigger picture

UBS reports that technology firms have allocated capital expenditures and research budgets on a scale comparable to entire industries, with 2024 spending alone exceeding the total research and development outlay of all publicly listed companies across Europe combined.

This wave of investment underscores the strategic urgency with which major players are seeking to secure leadership positions in the fast-developing AI market.

In the US, leading technology firms already account for a significant proportion of overall economic profit, giving them the financial capacity to sustain large-scale AI development initiatives.

Their robust balance sheets have allowed investment momentum to continue, even as doubts grow over the prospects for long-term returns.

The speculative foundations that create valuation risks

UBS highlights that the core challenge stems from the speculative basis of many AI use cases currently driving investment decisions.

Most revenue opportunities remain forecasted rather than realised, resulting in a widening gap between the vast sums being invested and the tangible returns generated from today’s operations.

Joe Tsai, Alibaba’s Cofounder, says to Business Insider: “I start to see the beginning of some kind of bubble... I start to get worried when people are building data centers on spec.”

Machine learning (ML) systems are widely expected to reshape sectors ranging from healthcare to financial services.

Yet many of these applications are still in experimental stages, with their commercial success reliant on technological advances and broad market uptake that may take years to unfold.

Despite this, current valuations across the technology sector already factor in ambitious assumptions about AI-driven growth, leaving limited room for error if projected cash flows fail to materialise.

This reflects the dynamics of the dotcom bubble, when internet firms achieved inflated valuations driven by projected rather than realised profitability.

As market realities fell short of those expectations, sharp price corrections inevitably ensued.

The uncertainties threatening returns

Several factors could influence whether AI investments achieve their expected returns over the medium term.

UBS finds that returns on capital investments in AI infrastructure remain uncertain, as companies continue to build data centres and acquire computing equipment without established revenue models to justify such expenditures.

Energy supply constraints add another layer of complexity to AI expansion plans.

The immense computational power required to train and operate AI systems consumes vast amounts of electricity, potentially creating infrastructure bottlenecks that could slow deployment and push operational costs beyond current forecasts.

This energy demand challenge not only threatens to limit the pace of AI rollout but also raises questions about sustainability and cost-efficiency in the long term.

Rising global competition adds further uncertainty to return calculations. 

How the change in AI competition could impact the AI bubble crisis

Regions with rapidly expanding research and development are building their own AI capabilities, which could narrow the competitive edges currently held by today’s technology leaders.

This rising competition threatens to erode profit margins and market share assumptions factored into existing valuations.

The timeline for broad AI adoption in traditional industries remains uncertain, as many promising applications face regulatory hurdles, technical challenges and institutional resistance that could push commercialisation beyond investor expectations.

Such uncertainties mean substantial portions of today’s AI investment may fail to deliver anticipated returns.

Effectively, companies are wagering on future technological breakthroughs and market uptake that remain largely unproven, while stock prices embed optimistic assumptions of success.

The risk is that when real-world outcomes confront these expectations, resulting market corrections could significantly reshape technology sector valuations and the broader economic forecasts hinging on AI-driven productivity gains across industries.

Lisa Su, AMD CEO, says to Business Insider: “The bubble talk is completely wrong. AI will fundamentally change everything over the next five years.”

Whereas, Thomas Siebel, CEO C3.ai, says: “There is absolutely an AI bubble and it’s huge. The market is way overvaluing some startups.”