Gartner says $3.5tn in AI infrastructure spending faces write-down risk by 2030
The research firm warns that planned capital expenditures exceed projected AI revenue, creating significant asset impairment risks for providers.
Key points
- Gartner forecasts more than $3.5 trillion in AI infrastructure spending could face impairment by the end of 2030.
- The firm predicts over 80% of neoclouds will cease independent operations by 2030 due to lack of distinct business models.
- AI data center owners would need $8.5 trillion in revenue to cover costs and financing, but only $3.6 trillion is forecast.
Gartner stated on 14 October that more than $3.5 trillion in artificial intelligence infrastructure spending faces potential impairment by the end of 2030. This assessment assumes that AI providers continue investing at their current accelerated pace without adjusting their capital expenditure strategies.
The research firm explains that companies must write down impaired assets when those assets can no longer generate sufficient earnings to recoup their initial costs. Senior director analyst Will Sommer told Bloomberg that while many providers aspire to become full-stack operators, market realities prevent all of them from achieving that status.
Financial Gaps and Revenue Projections
Gartner expects AI providers to invest nearly $1 trillion in infrastructure during 2026. The firm forecasts an additional $6 trillion in spending between 2027 and 2030. To cover these expenditures along with associated financing costs, AI data center owners would require $8.5 trillion in revenue through 2030.
This required revenue figure includes a 12% return on invested capital, which Gartner assumes represents the minimum hurdle rate for such investments. However, the firm projects that business and consumer AI revenue tied to data centers will not exceed $3.6 trillion during the same period.
Sommer noted that enterprise and consumer AI revenue appears insufficient to support profitable returns on all planned capital expenditures. If spending continues on its current trajectory, more than $4 trillion in asset value could face write-down risk through 2032. In this scenario, Gartner does not expect the 12% return to materialize until the end of 2032 at the earliest.
Impact on Neoclouds and Hyperscalers
The research note, authored by Sommer, Mark McDonald and John Lovelock, highlights a disparity between hyperscalers and neoclouds. Hyperscalers possess large, diversified revenue streams that could support AI spending for several years. Neoclouds, which rent out AI-optimized computing power, face greater pressure to earn returns quickly to remain operational.
Gartner predicts that more than 80% of neoclouds will stop operating independently by 2030. The firm attributes this consolidation to cost pressures and a lack of differentiated business models. Most AI-native providers operate asset-light businesses, meaning less than half of AI revenue flows to providers that own significant infrastructure.
Spending on AI-optimized cloud infrastructure is expected to grow from $66 billion in 2027 to $174 billion in 2030. Sommer argued this growth is insufficient. Most compute providers will not recover their investments before newer AI chip generations replace current hardware. Bloomberg named CoreWeave and Nebius as specialist providers facing higher exposure than Microsoft, Alphabet and Amazon. CoreWeave’s shares and bonds declined sharply this week amid concerns about its AI debt.
What should you do?
Security and technology professionals should monitor vendor financial health when evaluating long-term infrastructure partnerships. Organizations relying on neocloud services should assess the risk of provider consolidation or insolvency. Teams should review capital expenditure plans to ensure they align with realistic revenue forecasts.
What to do and how to stay safe: Gartner
- Review vendor financial stability and debt levels before committing to long-term AI infrastructure contracts.
- Diversify AI compute providers to reduce exposure to single-vendor failure or market consolidation.
- Monitor industry reports for updates on asset write-downs and neocloud operational changes.
- Align internal AI spending projections with conservative revenue estimates to avoid over-investment.
Step-by-step guide: Prompt Injection Warning Signs: How to Spot AI Manipulation
General security guidance from the Threat Times newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
How much AI infrastructure spending is at risk of write-down?
Gartner states that more than $3.5 trillion in spending faces possible impairment by the end of 2030.
Why are neoclouds predicted to stop operating independently?
Gartner cites cost pressure and a lack of a business model that distinguishes them from competitors.
What is the revenue gap for AI data centers?
Owners need $8.5 trillion to cover costs and returns, but Gartner forecasts only $3.6 trillion in revenue.
