AI Data Centers Are an Even Bigger Disaster Than Previously Thought
18 hours ago
- #AI
- #technology
- #finance
- Harris 'Kuppy' Kupperman, founder of Praetorian Capital, analyzed the financial challenges of AI data centers.
- AI data centers face rapid obsolescence and high replacement costs, making profitability difficult.
- Initial assumption of a 10-year depreciation for data center components was overly optimistic; actual lifespan is 3-10 years.
- Kupperman revised his break-even estimate for 2025 to $320-$480 billion in revenue, far exceeding current AI revenue of ~$20 billion.
- Scaling up data centers exacerbates financial issues, potentially turning an industry crisis into a national economic crisis.
- Industry professionals are equally baffled by the financial viability of AI data centers.
- Pessimists believe the AI sector is overhyped and a crash is inevitable.