Analyst says AI needs 250 million job cuts to justify $2.5 trillion spend
New analysis claims AI must replace 250 million roles to cover massive infrastructure costs, echoing dot-com bubble risks for providers.
Key points
- AI investment requires $2.5 trillion in annual revenue to break even.
- Current AI revenue estimates fall short by multiple trillions of dollars.
- Infrastructure providers face high depreciation costs from rapid hardware obsolescence.
Marin Ivezic, founder of Applied Quantum, states that artificial intelligence must replace 250 million full-time jobs to justify the current $2.5 trillion investment, according to TechRadar. The analysis highlights a significant gap between capital expenditure and potential returns for infrastructure providers.
Ivezic calculates that generating $2.5 trillion in annual revenue is necessary to cover costs. Current optimistic estimates place AI revenues between $150 billion and $250 billion per year. This creates a multi-trillion dollar shortfall that threatens the financial stability of companies building the underlying hardware and data centers.
The economics of the AI boom
The $2.5 trillion figure includes $1.5 trillion already spent on hardware since late 2022. An additional $1 trillion is projected for annual investments moving forward. Ivezic notes that this level of spending is unprecedented. The math relies on the assumption that each replaced job saves $100,000 in annual costs, requiring the elimination of 250 million positions to meet the revenue target.
That suggests the current business model for AI infrastructure is fundamentally unsustainable without massive labor displacement.
Hardware depreciation and pricing pressures
Infrastructure providers face a mismatch in asset lifespans. High-performance AI chips become obsolete every five years, while data center facilities last approximately 20 years. Ivezic calls this the "50-Cent Rule," requiring providers to generate 50 cents in annual revenue for every dollar invested to manage depreciation costs.
Meanwhile, the cost of using AI models has dropped significantly. GPT-4-class capability models are now 100 times cheaper than in 2022. However, infrastructure costs for providers have not decreased and are rising. This divergence makes it harder for providers to recoup their initial investments, a dynamic Ivezic compares to the dot-com bubble.
What should you do?
Security and technology leaders should monitor the financial health of cloud providers and AI infrastructure vendors. As hardware costs rise and model prices fall, vendor stability may become a concern. Organizations should diversify their AI service providers to mitigate risk. They should also review their own AI spending against tangible productivity gains to ensure their investments remain justified.
What to do and how to stay safe: Marin Ivezic
- Review current AI vendor contracts for clauses related to service continuity and financial stability.
- Audit internal AI spending to ensure it delivers measurable productivity gains rather than just experimental use.
- Diversify AI tooling across multiple providers to reduce dependency on a single infrastructure ecosystem.
- Monitor industry reports on hardware depreciation and data center costs to anticipate potential service price increases.
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 revenue is needed to justify AI investments?
According to the analysis, AI needs to generate approximately $2.5 trillion in annual revenue to justify the current investment levels.
Why do infrastructure providers face high costs?
Providers face high costs because AI chips become obsolete every five years while data centers last 20 years, creating large depreciation expenses.
What is the current estimated revenue from AI?
Optimistic estimates place current AI revenues between $150 billion and $250 billion per year, far below the required $2.5 trillion.
