Major global technology firms, including Microsoft, Alphabet, Amazon, Meta, and Oracle, are projected to spend nearly $700 billion on capital expenditures in 2026, primarily targeting infrastructure for artificial intelligence. While companies race to integrate AI, experts warn that the transition from initial experimentation to measurable productivity remains slow and often unquantified.
The Billion-Dollar Infrastructure Bet
cluster (priority): Deník.cz
The race for technological dominance in the artificial intelligence sector is driving unprecedented investment levels. According to a May 2026 report from the Center for Security and Emerging Technology, major industry leaders are aggressively expanding their computational power, often moving faster than their supply chains can accommodate. This massive capital allocation is not merely for software development but for the heavy industrial requirements of building and maintaining global data centers.
Analysts from the Columbia Business School have projected that the global construction of AI-ready data centers will require $8.2 trillion in investment over the next eight years. This figure represents approximately 2.8% of global GDP annually, highlighting the sheer scale of the shift toward AI-heavy infrastructure.
However, this massive spending creates a fundamental tension for investors. While many expect the high profit margins typical of traditional software—often reaching 80%—the reality of running data centers is closer to heavy industry or energy production. Electricity consumption and hardware maintenance costs suggest that margins are more likely to settle in the low double digits, as reported by CNN Prima NEWS.
Bridging the Gap Between Deployment and Productivity
AI’s $380 Billion Bet: Which Tech Giants Are Winning the Future?
While capital expenditures soar, the practical application of AI within businesses remains in an experimental phase. A recent survey of 155 Czech firms conducted by Deloitte found that while 73% of companies have an AI strategy or are currently preparing one, the actual measurement of AI’s impact is remarkably low.
According to Seznam Zprávy, only 7% of surveyed companies utilize formal processes and performance indicators to track what AI tools contribute to their operations. Nearly half of the firms—44%—do not track the usage of these tools at all.
Julie Slavík, an expert on technological innovation at Deloitte, suggests that many companies start their AI integration in reverse. Firms often deploy tools before defining specific goals, leading to a trial-and-error approach where benefits remain difficult to quantify. Furthermore, as noted by Marek Rehberger, general manager of Patria Corporate Finance, the integration period often involves a temporary dip in efficiency as teams learn to work with new technology. Even as efficiency gains begin to emerge, limitations persist.
AI is not yet able to evaluate the emotional and personal approach regarding the company being sold or bought, and it is a question of whether it will ever be capable of that. At the same time, AI analyses often do not go to the heart of the matter, but remain on the surface.Marek Rehberger, Patria Corporate Finance, via Seznam Zprávy
The Risk of Enshittification and Regulatory Hurdles
cluster (priority): Seznam Zprávy
As the industry moves past the initial honeymoon phase of AI—characterized by heavy subsidies and free or low-cost access—analysts are cautioning against a process known as enshittification. This term, coined by journalist Cory Doctorow, describes the cycle where platforms prioritize shareholder satisfaction and profit over user experience, often resulting in a degradation of product quality and an influx of sponsored content.
The challenge for European and Czech firms, in particular, is navigating this landscape while operating under stricter regulatory frameworks than their American counterparts. Marcel Červený of Bio Group noted that while American firms excel at marketing—often selling products before they are fully realized—European companies struggle with a regulatory environment that can act as a brake on innovation.
As reported by Deník.cz, the regulation of the European financial market makes it difficult to quickly deploy capital into new, promising projects. Despite this, there remains optimism that Europe can leverage its traditional strengths in quality engineering and applied technology if it can successfully bridge the gap between regulation and practical implementation.
Ultimately, the future of AI will likely be defined by a shift from the current phase of experimentation to a period where companies must justify their infrastructure spending with verifiable efficiency gains. Whether the market can sustain the current levels of investment without falling into the trap of platform degradation remains the central question for the remainder of 2026.
