The essentials in 30 seconds
- As of October 2026, exposure to an AI bubble is measured with 4 lenses: revenue, funding, customers and valuation.
- Companies that sell to AI builders gain the most while spending grows, and lose the most when it slows.
- The EQOS Asia AI study records valuations up ×4 in 18 months and about $200B a year of AI spending (April 2026).
- Seven groups fill the map: chips, Big Tech, OpenAI, neoclouds, Asian chipmakers, AI software and startups.
- This page is general information, not investment advice.
Which companies are most exposed to an AI bubble? The answer depends on where a company sits in the chain of AI spending, and on how that spending is paid for. This guide, part of the wider series on the AI bubble, gives you 4 lenses to measure exposure and a map of 7 company groups. It uses the EQOS Asia AI study as its base and reads as of October 2026.
On this page
Compare the EQOS Asia offers and request a 30-minute pre-diagnostic.
What is the short list of companies exposed to the AI bubble?
Exposure grows with dependence. A company is highly exposed when a large share of its revenue comes from AI spending by others, when it funds growth with borrowed money, and when a few customers buy most of its output. Companies that sell chips, rent computing capacity, or lend their name to model builders sit closest to the spending flow, so they feel a change in that flow first.
Companies that use AI to improve an existing business sit further away. Their core revenue continues if AI spending slows, and AI is an added bonus. The useful question is therefore a ratio: how much of this company’s value would remain if the AI part were removed? The sections below apply that ratio to 7 groups.
How do you measure a company’s exposure to an AI bubble?
Use 4 lenses in order. First, revenue dependence: the share of sales that comes from AI customers. Second, funding: whether growth is paid from operating cash flow or from debt and new share issues. Third, customer concentration: how many buyers represent most of the orders. Fourth, valuation: the price paid for each unit of earnings compared with the company’s own history.
Each lens gives a simple reading of low, medium or high. A company that scores high on 3 or 4 lenses has the most exposure to a fall in AI sentiment. The EQOS Asia AI study records valuations up ×4 in 18 months and about $200B a year invested in AI (April 2026), which shows why the valuation lens deserves a place in every reading.
How exposed are chip makers like Nvidia to an AI bubble?
Chip designers sit at the start of the spending chain. Every data centre order begins with accelerators, so their revenue tracks the budgets of a handful of buyers. Strong demand lifts earnings fast, and the same concentration means a pause in orders shows up quickly. Our page on Nvidia and the AI bubble applies the 4 lenses to the best-known name in this group.
The group also holds a real strength: its products are used in real workloads today, and its earnings are measurable. Exposure here is about the path of future orders, since current sales are real. Track order commentary, inventory levels at customers, and the share of revenue tied to the biggest buyers, which are public in company filings.
Why does Big Tech spending decide how big the AI bubble gets?
The largest cloud and platform companies are the biggest buyers of AI hardware, so their budgets set the pace for the whole chain. They also have large existing businesses that pay for the spending. Our review of Big Tech spending and the AI bubble compares how each of them funds its AI plans and what returns they report.
This group is the least exposed on the funding lens, because operating cash flow carries most of the cost. The exposure that remains is about returns: spending needs matching revenue from AI products. If revenue lags for several years, investors may ask for lower budgets, and that decision travels down to every supplier on this map.
Where do OpenAI and the neoclouds sit in the AI bubble chain?
Model builders sell AI directly, and they commit to large computing contracts before revenue arrives. The page on OpenAI and the AI bubble explains how revenue, commitments and outside funding connect. A private company has no share price, so you read its funding rounds and its contracts instead.
Neoclouds rent computing capacity and often borrow against the hardware they buy. Our guide to CoreWeave and the neoclouds shows why this leverage raises both the upside and the exposure. Both groups depend on customers who are still proving that AI revenue can match the cost of the computing.
How do Asian chipmakers, AI software and startups fit the AI bubble map?
Asian chipmakers manufacture and supply memory for AI hardware, so their order books follow the same spending cycle with a different customer mix. Read TSMC, SK Hynix and Asian chipmakers in an AI bubble for the details. For investors and companies in Singapore, ASEAN and Hong Kong, this group is the closest link to the physical supply chain.
AI software companies earn from usage and contracts, so their exposure is about valuation more than funding. The guide to AI software stocks and the AI bubble covers Palantir and peers. Small companies carry the highest risk of failure and the lowest cost of change, which the page on small AI startups and the bubble sets out.
How can you build your own AI bubble exposure map?
List the 7 groups, score each company you follow on the 4 lenses, and write the date next to the score. Rescore every quarter and note which lens moved. A company whose valuation lens rises while its revenue lens stays flat is a signal worth a closer look.
The EQOS Asia method page explains how the scoring logic works, and the pricing page lists the diagnostic options for a decision you need to make. This page is general information, not investment advice, and a score frames risk without predicting prices.
Frequently asked questions
Which companies are most exposed to an AI bubble?
Companies that depend most on AI spending by others carry the most exposure. Chip suppliers, computing-capacity renters and model builders sit closest to the spending flow, while companies that only use AI sit further away. As of October 2026 the 4 lenses of revenue, funding, customers and valuation give a clear first ranking.
Is Big Tech less exposed than a startup?
On funding, yes: operating cash flow pays most of the bill, so a slowdown reduces profit growth without threatening survival. A small startup depends on outside money and on a few customers. Both groups are exposed to the same question, which is whether AI revenue grows as fast as AI spending.
Does exposure mean a company will fall?
Exposure measures sensitivity. A highly exposed company gains most when spending grows and feels the largest effect when it slows. Prices depend on many other factors, so the score frames risk and leaves outcomes open. This page is general information, not investment advice.
How often should I update an exposure map?
Once a quarter fits, because companies publish results on that rhythm. Each quarter, check order commentary, funding changes and the share of revenue from the largest customers, then rescore the 4 lenses. Note the date, so the trend becomes visible over time.
Test your knowledge
1. Which lens measures how many buyers place most orders?
- Valuation
- Customer concentration
- Funding
Show the answer
Customer concentration.
2. Which group sits at the start of the spending chain?
- AI software
- Small startups
- Chip designers
Show the answer
Chip designers.
3. How many lenses does the page use to measure exposure?
- 4
- 7
- 10
Show the answer
4.
4. Why is Big Tech the least exposed on funding?
- Operating cash flow pays most of the cost
- Governments pay the bill
- It has no AI spending
Show the answer
Operating cash flow pays most of the cost.
5. What does a company’s private status change for a reader?
- Nothing, shares are always listed
- You read funding rounds and contracts instead of a share price
- Exposure drops to zero
Show the answer
You read funding rounds and contracts instead of a share price.
About the author
Stéphane Jambu
President, EQOS Forecast Asia
Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile
Sources: EQOS Asia AI impact study (eqos-asia.com/case-studies/ia-2026.html), EQOS Asia pricing page (eqos-asia.com/pricing.html) and method page (eqos-asia.com/method.html), consulted on 22 October 2026. Page updated on 22 October 2026.