The AI Bubble: Inflation, Burst, Opportunity
Home / Thought Leadership / Blog
AI INSIGHTS | NOVEMBER 2026
The AI Bubble: Inflation, Burst, Opportunity
PART 1
The AI bubble: how a story became a loop
The bubble began with a narrative. AI labs presented a compelling vision: remarkable productivity, tasks transformed, and wide profit margins for the organizations that adopted early. Belief in that story drove investment, first from institutions and then from the public.
Delivering on that promise required enormous computing capacity, which created a loop. Investors fund the AI labs. The labs commit to cloud providers. The cloud providers rent or build data centers and place vast hardware orders with chipmakers such as Nvidia. The chipmakers' growth then validates the original investment, and the cycle repeats at a higher valuation.
|
Investors |
→ |
AI labs |
→ |
Cloud providers |
→ |
Chipmakers (Nvidia) |
|
↻ Chip demand and rising valuations flow back to investors, and the loop restarts |
The difficulty is that results have not kept pace with the spending. The evidence is consistent across independent sources:
|
~95% of enterprise generative-AI pilots in the study showed no measurable P&L impact. MIT NANDA, 2025 |
|
9 in 10 vs 37% use AI in at least one function, yet only 37% report any enterprise-level EBIT impact. McKinsey, 2026 |
|
10-20-70 10% algorithms, 20% technology and data, 70% people and process. BCG |
The BCG principle explains the pattern. AI transformation is only about 30% a technology challenge. Roughly 70% is people, process and organizational change. Yet the world rushed to build agents, and most of that effort stalled at the proof-of-concept stage. Almost every organization now has its own POC graveyard: pilots that impressed in a demonstration and never reached production.
That gap between promise and proof is the pressure the market could no longer ignore.
PART 2
The bubble bursts: three forces converge
No single event bursts a bubble. Three forces came together.
|
1 |
Policy risk. Democrats won both chambers of Congress in the 3 November midterms, and their agenda includes limits on data-centre power costs and mandatory AI system testing. The prospect of tighter regulation is real, although a presidential veto and the need for bipartisan support may slow legislation. |
|
2 |
Missing returns. Investors began to price in what the research already showed: adoption is broad, but financial impact is narrow. |
|
3 |
Market losses. About $3.2 trillion in market value was wiped out in the sell-off that followed the midterms, and the Nasdaq Composite fell 5.56% in a single day. |
These are not three separate stories. Policy risk exposed the returns gap, and the returns gap made the sell-off severe. The natural question is what happens next.
PART 3
After the burst: why this is an opportunity
History offers a guide. When steam engines first powered locomotives, they consumed large amounts of expensive coal. Engineers then redesigned the engines and the process around them, and each journey required far less coal. A reasonable expectation was that coal demand would fall. The opposite happened. Because travel became so much cheaper, many more people and businesses began to use it, and total coal consumption rose sharply.
Economists call this the Jevons paradox: when efficiency lowers the effective cost of using a resource, total consumption of it rises.
|
The same dynamic is now arriving for AI. As the cost of using AI falls, organizations that could never justify it will adopt it. Every department and every routine action will be built on AI, and demand for AI solutions will rise dramatically. |
Organizations must be ready to meet that demand, for their customers and internally. The tools and habits that produced the POC graveyard will not be enough. Handing someone a car with a full tank does not help if they do not know how to drive. What is needed is a mindset and an approach for building meaningful, economical AI solutions. That capability is what it means to be AI native.
Scaled Agile offers three AI-native programs, each building a different capability:
|
Program |
Duration |
Purpose |
|
AI Native Foundations |
2 days |
A hands-on workshop that builds AI-native fluency for everyone in the organization. |
|
AI Native Value Architect |
2 days |
The most critical capability to build: a systematic approach to architecting value across the AI development lifecycle. |
|
Leading the AI Native Organization |
1 day |
The leadership capability the AI era demands most. |
|
FOR LEADERSHIP TEAMS SAIL: Scaling AI for Leadership A 3-day, hands-on agentic AI workshop for leadership teams who want to move from awareness to execution. |
If an organization is serious about succeeding in the AI era, now is the time to lay the right foundation.
It is not what you eat that kills you. It is what eats you kills you.
The failure of tool-led adoption is what may be eating your organization today. Become AI native, and you will be well placed to succeed in what follows the bubble.
Take the next step in your career journey
Unlock your full potential with Gladwell's expert career Coaches.
Connect with us today and discover how we can help you grow.