Paolo Ardoino, CEO of the world’s leading stablecoin issuer, provided a thorough analysis of Big Tech’s spending in AI infrastructure on July 4. He emphasized significant structural risks arising from the current wave of capital investments. As companies navigate this landscape, they are investing vast amounts in data centers, GPUs, and power capacity, all while profits seem frustratingly elusive amid growing competition from open-source alternatives.
Ardoino highlighted four economic mismatches that should raise alarms in the AI sector. The first concern is that the prices of compute tokens do not truly reflect their underlying costs. This is compounded by the substantial initial investments required versus the delayed timeline to achieve profitability. Additionally, the capital maturity schedules for financing often mismatch the hardware lifecycles, with AI chips depreciating in just 3 to 5 years, while financing structures expect much longer return periods. Lastly, open-source AI models are increasingly undermining the commercial revenues expected to support this high level of investment.
The financial projections surrounding AI spending are staggering. JPMorgan has forecast that global spending related to AI could reach $5.5 trillion by 2030, a number that climbed sharply according to revised estimates in June 2026. Furthermore, Goldman Sachs anticipates that major firms like Microsoft, Meta, Amazon, and Alphabet will collectively spend approximately $5.3 trillion from 2025 to 2030. Hyperscaler capital expenditures are also set to rise sharply, expected to jump from $650 billion in 2026 to an astonishing $1.1 trillion in 2027, with major players aiming for a 77% increase in annual spending.
Despite these colossal expenditures, recent data from the Bureau of Economic Analysis indicates only a modest 1.5% growth in the information sector in Q1 2026. Corporations such as Amazon and Uber are already expressing concerns over the escalating costs linked to AI, which indicates a brewing tension in the market.
Ardoino's concerns are not new. Previously, he warned that an AI bubble could pose significant risks to Bitcoin. The rationale is straightforward: a sharp correction in AI-related stocks could potentially lead to a broader market fallout affecting correlated assets, including cryptocurrencies. Portfolios containing both Nvidia and Bitcoin might face margin calls, prompting widespread selling activity.
Tether has not remained passive in this environment. The organization has been investing in AI infrastructure while promoting decentralized solutions through initiatives like QVAC, positioning itself strategically in the confluence of stablecoins and decentralized AI.
For investors in the cryptocurrency space, monitoring a couple of key signals is crucial. First, it will be important to observe whether the earnings of hyperscalers begin to exhibit declining returns from their AI capital expenditures. The second point of interest is tracking the adoption rate of open-source models and their potential to significantly undermine the pricing of commercial AI. The anticipated 77% surge in planned 2026 spending creates a narrow window for error should revenue growth fail to keep pace.