The Euphoria for Artificial Intelligence Collides with the Reality of Its Numbers

• Capital spending on AI could reach $760 billion by 2026.
• Debt linked to artificial intelligence projects has reached $489 billion.

The narrative of unlimited growth surrounding artificial intelligence is beginning to face financial pressure. An analysis published on July 30, 2026, by analyst Bret Jensen indicates that major tech companies have raised their investments in infrastructure to record levels, while some companies are showing lower cash generation and investors are starting to demand greater proof of profitability.

For more than three years, since the launch of ChatGPT at the end of 2022, companies like Microsoft, Alphabet, Amazon, Meta, and Oracle have driven an accelerated expansion of data centers, chips, and computing capacity to compete in generative artificial intelligence. By 2026, the combined capital spending of the leading providers could reach between $700 billion and $760 billion, compared to about $410 billion in 2025.

Alphabet was one of the most significant signals of this new phase, as the company reported a negative free cash flow of $5.9 billion during a quarter, a fact that had not occurred since its IPO in 2004. Additionally, it raised its annual capital expenditure forecast by $15 billion, to a range of between $195 billion and $205 billion.

But the figures also reflect tensions in other companies in the sector. Meta recorded a 91% year-on-year drop in its free cash flow after accelerating its investments in artificial intelligence. Oracle closed its last fiscal year with a negative free cash flow of $23.7 billion and expects to allocate around $40 billion more to capital spending during its new fiscal period.

Tesla also showed signs of financial pressure related to its artificial intelligence strategy. The company recorded negative free cash flow for the first time in two years while increasing its investments in AI-driven robotics, with projected capital spending of at least $25 billion during 2026.

This adjustment was also reflected in assets related to technological infrastructure. For example, the Philadelphia semiconductor index accumulated a rise of over 90% during the first half of 2026, but subsequently fell nearly 25% from its all-time highs reached on June 22, before recovering some of the lost ground.

Financing for the sector is another point under observation. According to estimates from Goldman Sachs, debt issuances linked to artificial intelligence projects reached $489 billion by mid-July 2026, surpassing the $321 billion issued throughout 2025.

In addition to the increase in debt, the ecosystem is beginning to use more complex financing structures among its main players. Nvidia is evaluating backing the construction of an AI data center complex for OpenAI with a capacity of 10 gigawatts, a project whose potential value could reach $250 billion. The possibility of large companies financing part of the infrastructure for their own clients has raised questions about interdependence among companies in the sector.

Doubts also extend to the future IPOs of artificial intelligence companies. As reported by CriptoNoticias, SpaceX lost more than $1.2 trillion in market value since its post-IPO peak, a drop that could reduce investor appetite for new large public offerings like those planned by OpenAI and Anthropic.

This scenario is compounded by competition from models developed by Chinese companies like DeepSeek and Moonshot, which have increased pressure on current leaders by offering alternatives with lower development and operational costs. This forces U.S. companies to demonstrate that their investments in infrastructure can translate into sustainable advantages.

Artificial intelligence thus enters a phase where economic efficiency begins to weigh more than investment capacity. The next stage of the market could be marked by a stricter selection among projects, favoring companies capable of turning infrastructure into profitable products and use cases with proven demand.

For users and companies adopting these tools, the shift could translate into an evolution towards solutions more focused on productivity and economic return. The competition to reduce costs could also favor more efficient models and expand access to artificial intelligence services, although the pace of expansion will increasingly depend on the sector's ability to justify its enormous investments.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.