Major concerns are emerging about artificial intelligence spending practices as parallels to the late 1990s dot-com bubble surface. According to recent analyses from credit rating agencies and financial institutions, artificial intelligence investment patterns increasingly mirror the excessive speculation that preceded the tech crash of 2000. Big tech companies are pouring unprecedented capital into AI infrastructure while using accounting methods that obscure the true scale of financial obligations.
Moody’s analysis, as reported by Fortune, reveals that the five largest U.S. cloud computing companies have accumulated $662 billion in future data center lease commitments that currently remain off their balance sheets. These companies—Amazon, Meta, Alphabet, Microsoft, and Oracle—collectively face $959 billion in total future lease commitments for data centers that have not yet been constructed.
Artificial Intelligence Spending Reveals Hidden Liabilities
The $662 billion figure represents lease agreements that have not yet commenced but will eventually appear as liabilities once services begin. According to the Moody’s assessment, this amount equals 113% of these companies’ most recent adjusted debt, effectively doubling their apparent debt obligations. Additionally, even if companies cancel or fail to renew leases, they remain contractually bound to pay significant residual value guarantees.
This accounting treatment, while permissible under U.S. Generally Accepted Accounting Principles, raises questions about transparency in AI infrastructure investments. The practice allows companies to delay reporting massive financial obligations until lease agreements become active, potentially obscuring the true financial picture from investors and stakeholders.
Capital Expenditure Concerns Mount
Investment analysts are increasingly worried about the scale of AI capital expense spending and companies’ reliance on debt financing. UBS data indicates that aggregated capital expenditure among AI hyperscalers could reach $770 billion in 2026, according to CNBC reporting. This projection represents a 23% increase over previous expectations following announcements from Amazon, Meta, and Alphabet about expanded spending plans.
UBS credit strategists estimate these increases will necessitate $40 billion to $50 billion in additional borrowing, potentially driving public market debt issuance to between $230 billion and $240 billion this year. The shift toward debt financing represents a departure from previous assurances that AI investments would be funded through generated cash flow.
Al Cattermole, fixed income portfolio manager at Mirabaud Asset Management, told CNBC that the change raises fundamental questions about creditworthiness. He noted that for years, investors were assured that AI spending constituted equity risk funded by cash flow rather than debt obligations.
Venture Capital Behavior Signals Uncertainty
Meanwhile, venture capital investment patterns in artificial intelligence startups suggest growing uncertainty about market winners. TechCrunch reported that investor loyalty in the AI sector has largely disappeared, with venture capital firms increasingly diversifying investments across competing companies rather than concentrating resources on specific partnerships.
The Wall Street Journal highlighted emerging businesses called “neolabs” that focus exclusively on long-term AI research without immediate profit expectations. Some of these ventures are securing billions in funding from venture capital firms despite having no products or revenue, echoing investment patterns from the late 1990s technology bubble.
However, some AI applications are demonstrating practical value. Multiple software companies report successfully using large language models to parse human requests and facilitate system integration, though typically in conjunction with other complementary technologies for pattern recognition and calculations.
Historical Parallels Emerge
The current situation evokes memories of the late 1990s and early 2000s, when speculation about internet and mobile computing technologies drove massive investments in unprofitable ventures. That era prioritized “eyeballs” and attention over traditional business metrics, ultimately resulting in widespread company failures and destroyed shareholder value.
In contrast to those failures, some companies like Amazon successfully pursued long-term strategies involving sustained spending beyond revenue for years. Nevertheless, most ventures from that period collapsed, with many going public through investor-pushed exits before financial realities became apparent.
The combination of off-balance-sheet liabilities, unprecedented capital expenditure levels, debt-financed AI infrastructure, and diversified venture capital strategies suggests potential structural vulnerabilities in the current artificial intelligence investment landscape. While no definitive proof of systemic problems exists, the pattern of warning signs continues to accumulate as spending accelerates and scrutiny of financial practices intensifies.
