The Risks of AI Data Center Investments
In a recent opinion shared on X, Tether CEO Paolo Ardoino highlighted concerns about the sustainability of Big Tech's investments in AI data centers. He pointed out that these efforts rely heavily on subsidized computing and hardware that depreciate rapidly within three to five years. According to Ardoino, four structural issues are currently putting the sector at risk.
The warning comes as hyperscalers continue to pour significant resources into infrastructure, despite not yet seeing clear returns on their investments.
The Four Key Issues Identified
Ardoino outlined four main problems facing AI companies:
- Token prices do not match costs: The cost of using AI services is often higher than the revenue generated from token sales.
- Profitability timelines do not align with investments: Companies are investing heavily in infrastructure without a clear path to profitability.
- The maturity of capital does not match asset life: The financial resources available for investment do not correspond with the lifespan of the assets being developed.
- Open-source AI could lower revenues: The rise of open-source AI models may reduce the revenue streams of proprietary solutions.
Soaring Investment Figures
The scale of these investments is staggering and continues to grow. In its midyear outlook released on June 24, JPMorgan increased its estimate for global AI-related capital spending through 2030 to $5.5 trillion, up from $5.1 trillion. The bank also predicts that AI-related debt financing will reach $4.1 trillion.
JPMorgan forecasts that hyperscaler capital spending will hit $650 billion this year and exceed $1.1 trillion by 2027. Microsoft alone plans to spend approximately $190 billion in 2026, marking a 61% increase from the previous year.
Goldman Sachs estimates that Meta, Microsoft, Amazon, and Alphabet will collectively spend $5.3 trillion on capital expenses between 2025 and 2030. This year, these four companies plan to spend $725 billion, which represents a 77% increase compared to last year’s $410 billion.
Alphabet recently raised $84.75 billion for AI infrastructure, described as the largest US equity capital raise ever, according to reports.
No Clear Returns Yet
Ardoino's concerns about profitability reflect a broader uncertainty regarding whether these massive investments will eventually pay off. On average, companies are spending $11.5 million on AI this year, but most cannot demonstrate any clear return on investment. Data from the Bureau of Economic Analysis shows that growth in the Information sector slowed to 1.5% in the first quarter of 2026, down from 3.2% in the third quarter of 2025.
Ardoino's warning about open-source AI reducing revenues aligns with a trend that has been developing over several months. Companies that previously encouraged staff to maximize AI usage (also known as “tokenmaxxing”) are now scaling back as CFOs question rising API costs.
Examples of this shift include:
- Amazon removing its internal leaderboard tracking employee AI use.
- Uber exhausting its 2026 AI coding budget in just four months and setting a $1,500 monthly cap per employee.
- Meta cautioning 6,000 staff about rapidly increasing costs.
IDC projects that by 2028, 70% of leading AI adopters will use multiple models instead of relying on a single vendor, potentially sparking a price war.
Regulatory Concerns
Regulators are also expressing concerns. The Bank for International Settlements warned in its annual report that a sharp drop in AI investment could negatively impact global stock markets more than past recessions. The BIS identified AI as one of three main risks to the economy.
Zhang Tao, the BIS chief representative for Asia and the Pacific, stated, “the speed of a correction could be much faster than previous banking crisis episodes.”
Differing Perspectives
Not everyone shares Ardoino’s pessimism. Wedbush analyst Dan Ives described the current AI buildout as “an arms race” that no major company can afford to ignore. He believes the sector will start generating profits in the next six to twelve months.
JPMorgan also expects strong profits, predicting that operating cash flow will exceed $900 billion by 2027.
Thomas Hayes, chair of Great Hill Capital, offered a more balanced view, suggesting that one or more major companies may announce reduced capital spending in upcoming earnings reports. For now, the upcoming earnings season will be crucial. If any of the big spenders cut back, as Hayes predicts, it will serve as the first real test of the issues Ardoino highlighted.
If you're reading this, you’re already ahead.
Stay there with our newsletter.
No comments:
Post a Comment