AI valuations up 4x in 18 months: is that an AI bubble?
The essentials in 30 seconds
- A ×4 rise in AI valuations over 18 months is a bubble signal, and earnings growth decides whether it is a bubble.
- The path equals about 8% growth per month, or roughly 152% per year.
- The EQOS Asia AI study scores the enterprise AI market at 0.337, in the fragile zone (April 2026).
- Divide price growth by earnings growth: a result near 1 means the rise is earned.
- 2 guides below cover price to earnings and market value against GDP, as of October 2026.
AI valuations up 4x in 18 months is the headline number of the AI stock bubble debate. This page turns that number into rates, multiples, and checks, as of October 2026. The EQOS Asia AI study records the ×4 rise and a φ(S) score of 0.337 for the enterprise AI market. It is one of the AI bubble signals that decision makers track, and it ends with a method you can repeat. This page is general information, not investment advice.
On this page
Compare the EQOS Asia offers and request a 30-minute pre-diagnostic.
What does an AI valuation rise of ×4 in 18 months mean per year?
A valuation that quadruples in 18 months compounds at about 8% per month. Over a full year the same path equals a rise of roughly 152%. These rates come from simple arithmetic on the ×4 figure in the EQOS Asia AI study (April 2026), and you can check them with a calculator. Rates make the claim concrete, because few businesses grow revenue at 152% a year for long.
The rate also shows how sensitive the story is to the start date. Measure from a market low and the rise looks steeper, measure from a calmer month and it looks gentler. Write down the start date, the end date, and the index used before you compare 2 claims. A ×4 figure without those 3 details is a headline, and a ×4 figure with them is a measurement.
How much must earnings grow to support an AI valuation rise of ×4?
Prices follow earnings in the long run, so a ×4 rise asks for a matching rise in what companies earn. If earnings double while prices quadruple, buyers pay twice as much for each dollar of profit as they did 18 months ago. That gap between price growth and earnings growth is the core of the AI stock bubble question, and it is measurable each quarter.
Take 3 inputs from public filings: the price change, the earnings change, and the date range. Divide the first by the second to see the multiple expansion. A result near 1 means the rise is earned. A result of 2 or more means expectations carry half the move, and expectations can change faster than profits do.
How do AI valuations split into earnings growth and higher multiples?
Every valuation rise splits into 2 parts: earnings growth and a higher multiple. Earnings growth is backed by customers who pay. A higher multiple is backed by the hope that growth will continue. Sorting the ×4 move into these parts tells you how much of the rise a customer paid for and how much an investor imagined.
The multiple that most readers know is price to earnings. The guide on AI stock price-to-earnings ratios shows how to read it for AI companies, including the cases where earnings are still small and the ratio looks extreme. Read the ratio together with the growth rate, because a high ratio on fast growth means something different from a high ratio on flat sales.
How large can AI valuations grow before the economy has to carry them?
Valuations also have a ceiling set by the size of the economy. The total value of AI companies, compared with national output, shows how much of the world’s income the market expects AI to earn. The guide on market capitalisation against GDP puts that comparison in order and shows the scale of the claim.
Size matters because a ×4 rise starts from a larger base each month. Growth from a small base is easy to repeat, while growth from a large base needs ever more revenue. When the base reaches a visible share of GDP, further rises need new customers, new budgets, and new pricing power, and each of those takes years to build.
What does the EQOS Asia study add to the AI valuations question?
The EQOS Asia AI study places the ×4 rise next to the other numbers that matter: about $200 billion a year invested, 95% of AI projects failing in production (MIT 2025, cited by the study), and a φ(S) score of 0.337 for the enterprise AI market. The score sits in the fragile zone, where the system works and a shock can tip it.
The study calls its verdict bifurcation: a soft landing and a break both stay possible. For valuations, that means the ×4 rise is neither proof of a bubble nor proof of a boom. It is one input, and the EQOS Asia method page shows how inputs combine into a score that anyone can recompute.
How can you test AI valuations after a fast rise?
Run a 4-step check on any AI stock that rose fast. Record the price change, record the earnings change, divide one by the other, and compare the result with the same figure for the broad market. Do this each quarter with the same index and the same dates, so the numbers line up. A written record turns an impression into a trend.
Add a funding check. Ask who bought the shares and who paid for the growth: customers through revenue, or new investors through fresh capital. Rising revenue per customer supports the rise, while rising share issuance alone points to a market that needs new money to keep moving. The Λ simulator lets you score your own inputs in 10 minutes.
What should a decision maker do about AI valuations this quarter?
Size AI exposure to what your business or portfolio can carry if the ×4 rise reverses half of its path. Stage large commitments over several quarters, set review dates, and write the thresholds in advance. This page focuses on method and leaves out share prices and forecasts. As of October 2026 the useful output is a process that you can repeat.
If a decision of yours depends on AI valuations in Singapore, ASEAN, or Hong Kong, a 30-minute pre-diagnostic with EQOS Asia scores your case on the same method. You bring the decision and the inputs, and the session returns a score and a ranked set of options.
See also in this guide
Frequently asked questions
What does AI valuations up 4x in 18 months mean?
It means the combined value of AI companies in the EQOS Asia AI study multiplied by 4 over 18 months. That path equals about 8% growth per month, or roughly 152% per year. The figure describes prices, so you read it next to earnings and revenue before drawing a conclusion about a bubble.
Is a 4x rise in AI valuations proof of a bubble?
A rise alone confirms nothing. A bubble shows when prices outrun earnings and the growth needs new investors to continue. A ×4 rise with earnings that also grow strongly describes a boom. The test is the ratio of price growth to earnings growth, which you can calculate each quarter from public filings.
Which data shows whether AI valuations are too high?
Three data sets help: price to earnings, market value against national output, and revenue growth. Each one comes from public sources and updates quarterly. Together they show whether prices rest on profits, on the size of the economy, or on hope. The 2 guides under this page explain the first 2 sets.
How often should I recheck AI valuations?
Once a quarter suits most decisions, because earnings arrive each quarter. Use the same index, the same dates, and the same formula, then compare with last quarter. A multiple that expands for 2 quarters in a row signals rising risk, and a multiple that holds signals that earnings are catching up.
Test your knowledge
1. What monthly growth rate does a ×4 rise over 18 months equal?
- About 2%
- About 8%
- About 25%
Show the answer
About 8%.
2. What does a price-to-earnings gap of 2 or more mean in this page?
- Expectations carry half the move
- Earnings carry the whole move
- Prices have fallen
Show the answer
Expectations carry half the move.
3. Which score does the EQOS Asia AI study give the enterprise AI market?
- 0.618
- 0.095
- 0.337
Show the answer
0.337.
4. What does the funding check ask?
- Which index is largest
- Who paid for the growth
- How many analysts cover the stock
Show the answer
Who paid for the growth.
5. How often does the page suggest rechecking valuations?
- Every day
- Once a quarter
- Once a decade
Show the answer
Once a quarter.
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 method page (eqos-asia.com/method.html) and contact page (eqos-asia.com/contact.html), consulted on 22 October 2026. Page updated on 22 October 2026.