Is There an AI Bubble? The Boom, the Risks and What to Watch

Is the AI Boom a Bubble?

Nobody can say for certain, and the evidence points both ways. The companies leading the AI build-out are profitable and report strong demand, but they are also spending hundreds of billions of dollars on infrastructure, and the market is increasingly asking when that spending will pay off. This page tracks the debate each week, with the numbers behind it.

Latest Update: Week of 2-6 November 2026

• The post-election rally reversed. Markets surged straight after the midterms, then turned sharply on Thursday as investors weighed whether sweeping Democratic wins could mean tighter limits on data-centre construction and weaker AI spending. The Nasdaq shed more than 1,400 points and entered a correction, a fall of at least 10% from its previous peak. Nvidia and Broadcom each fell roughly 9%, Oracle about 7% and CoreWeave about 11%. Stocks rebound after tech selloff that pushed Nasdaq into correction.
• Data-centre opposition is local. Wall Street analysts said resistance is more likely to surface at state and local level, and voters have approved more than 40 ballot measures restricting data centres across the country. See our guide to the midterm results.
• Valuations, not earnings. Bank of America said the recent sell-off in AI-exposed hardware came from lower valuation multiples, not lower earnings estimates. 
• A partial rebound on Friday. AI stocks recovered some of the losses, with Nvidia up 2.8% after drops of 9.3% and 3.2% in the previous two days. The Nasdaq Composite closed 1.09% higher at 24,579.42. AMD rose 11% and the semiconductor index 4.2%. One report said the Nasdaq was still on course for a fourth straight weekly fall. AI stocks halt their brutal, worldwide sell-off as Wall Street trims its loss for the week.
• Rates are in play. Strong US jobs data raised expectations of a December Fed rate hike, and the Bank of England raised Bank Rate to 4% on 5 November. See our Federal Reserve forecast and Bank of England analysis.
• The spending bill is still huge. Amazon, Alphabet, Microsoft, Meta and SpaceX are set to spend up to $828 billion this year, and four of the five had negative free cash flow in the third quarter. Combined buybacks at the main hyperscalers fell to $4.4 billion in the quarter, the lowest since 2017.
• Demand evidence continues. AWS revenue grew 39.3% last quarter, with a backlog of $578 billion. 
• Next: Nvidia reports on Tuesday 17 November after the US close, with consensus expectations of about $108.8 billion in revenue and EPS of $2.47. The Fed meets in December, and markets will watch hyperscalers’ 2027 spending plans. 

What Is a Market Bubble?

A bubble forms when asset prices rise far beyond what the underlying fundamentals can justify, driven by expectations of ever-higher prices rather than by earnings or cash flow. Bubbles are usually obvious only in hindsight. The dot-com crash of 2000 is the standard comparison for AI.

The Bear Case

• Spending is unprecedented. Analysis by Apollo puts US AI spending at about 2% of GDP, against 1.2% at the peak of the telecom and fibre build-out in 2000, and heading towards 3%. 
• Returns haven’t caught up. Capex is growing faster than the revenue it generates, and free cash flow has turned negative at several of the biggest spenders.
• Shareholder returns are being cut. The collapse in buybacks shows money being redirected from investors to infrastructure.
• Weaker links. One prominent tech investor quoted by Benzinga warned that “neocloud” providers have leveraged balance sheets and heavy negative cash flows, unlike the cash-generating hyperscalers. 
• Concentration and rates. A handful of stocks carry the market, and rising rates raise the discount applied to future profits.
• Policy risk. Data-centre approvals, power supply and local opposition could slow the build-out.

The Bull Case

• The leaders are profitable. XTB analysts argued in 2025 that today’s tech giants are delivering real profits and trade at far lower valuations than internet stocks did before the dot-com crash. 
• Demand is visible. Revenue is growing rapidly at AI companies such as Anthropic and OpenAI, and cloud backlogs run into hundreds of billions of dollars. 
• Cash-rich spenders. The biggest investors fund much of the spending from established businesses, which gives them more room to adjust if demand slows.
• Possible upside from a slowdown. If spending peaks, free cash flow at the big hyperscalers could improve.
• The sell-off may reflect valuation, not earnings. Bank of America’s view is that earnings estimates haven’t fallen, even though share prices have.

Read our guide to Bull Market vs Bear Market in understanding the difference between them.

Key Indicators to Watch

Then and Now: How Does It Compare With the Dot-Com Era?

Simplified comparison. Past episodes are not a reliable guide to what happens next.

Company Deep-Dives

• AMD vs Broadcom: two AI chip approaches compared
• Oracle: the RPO backlog and the cash-flow debate
• AI ETFs and our guide to AI and the future of technology

How Can UK Investors Approach the AI Debate?

• Avoid all-or-nothing bets. If you hold AI stocks, check how much of your portfolio they already represent, including through index funds, which are heavily weighted to them.
• Diversify. Broad ETFs spread exposure across sectors, and Investment Plans automate regular, diversified investing, which reduces the impact of timing.
• Think about time horizon. Volatility around the AI theme can be large, and short-term moves are hard to predict.
• Be careful with leverage. CFDs magnify losses as well as gains. Read our guides to stock CFDs vs buying shares and CFD trading strategies.
• Consider an ISA. Shares and ETFs can be held in a Stocks & Shares ISA.

Tax treatment depends on your individual circumstances and ISA regulations which may change.

Risk Warning

CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. 78% of retail investor accounts lose money when trading CFDs with this provider. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money.

Capital at risk. Investment values can rise or fall. Tax treatment depends on your individual circumstances and ISA regulations which may change.

This article is for informational and educational purposes only. It does not constitute financial advice or a personal recommendation. Past performance is not a reliable indicator of future results.