2026’s AI market is projected to experience its biggest reckoning with the soaring capital expenditures (CAPEX), volatile stock movements, and investor skepticism that could shatter the perception of universal winners. AI CAPEX will split between cash-burning AI spenders and profitable AI infrastructure.
2026 will come and bring along a strong differentiation birthed from 2025’s turbulent fourth quarter of tech selloffs and rallies.
Investors, who have pumped venture capital (VC) with $176.5 billion so private startups, such as OpenAI, are now scrutinizing the free cash flow of Big Tech giants, such as Amazon and Meta.
In 2025, Big Tech giants morphed their companies to become hyperscalers to fund their AI strategies, even if that meant turning their debt markets into a financing project for (GPU) and data center buildouts.
If incremental AI revenues fail to outpace massive AI investment bubble concerns related to expenses, then margins will compress and the performance gap will widen, according to CNBC.
Nvidia, among other infrastructure providers, is among the clearest beneficiaries of the spending boom that will spill into 2026.
Big Tech Spending on AI
AI’s new autonomous identity has widened the advanced technology’s doors to welcome bigger capital toward a certain group of companies. The same identity that controls computing power and scale, triggering the AI CAPEX boom driven by the high cost of advanced AI agents.
“Every company seems to be winning,” he said. “It’s very important to differentiate between different types of companies, which is what the market might start to do,” Stephen Yiu, chief investment officer at Blue Whale Growth Fund, told CNBC.
For Yiu, retail investors, especially those purchasing AI centered Exchange-Traded Funds (ETFs), usually fail to differentiate startups with no clear business model, companies burning cash on AI infrastructure, and firms that profit by supplying chips, servers, and tools.
It’s a behavior that helps inflate the AI CAPEX bubble, where spending size matters more than financial results.
The market is divided between private AI startups and public companies spending heavily on AI and infrastructure. According to PitchBook data, private firms such as OpenAI and Anthropic raised $176.5 billion in 2025’s first three quarters, driving expectations of an AI CAPEX boom.
Meanwhile, Big Tech AI spending is driven by platform giants channeling billions into chips, servers, and energy intensive systems. Most Magnificent 7 stocks are trading at a premium as AI CAPEX rises rapidly, according to Yiu.
“The AI “froth” is concentrated in specific segments rather than across the broader market,” Julien Lafargue, chief market strategist at Barclays Private Bank, tells CNBC.
Therefore, AI investment bubble concerns are the greater in companies with limited revenue.
Sustainable AI Investments
The challenge is not limited to companies, as countries with capital, land, and power are better positioned to meet growing AI infrastructure spending projections, while others risk falling behind. In fact, they’re most suitable to meet these projections.
Google and Meta for example have shifted from software focused models to asset heavy operations, like investing billions in GPUs. This has intensified analysis of Big Tech spending on AI, especially as data centers become long-term costs.
Dorian Carrell, head of multi-asset income management at Schroders, believes that old methods for valuing major tech firms are non-effective nowadays, and investors must rethink valuations as spending grows.
Carrell questioned whether current prices make sense given rising AI CAPEX demands. On the other hand, debt markets help fund expansion; however, they are not really a reliable option for all firms. Smaller companies face scaling enterprise AI ROI challenges, as turning AI spending into profit remains difficult.
Additionally, infrastructure depreciation is a pressure factor. Yiu warned that incoming earnings will reflect the costs and raise concerns about the optimism of AI infrastructure spending projections.
Now sustainable AI investments are seen as the ultimate decision, where they focus on stable growth rather than scale at any cost.
Projects related to AI data center investment are already changing where AI development happens, favoring regions with reliable power and funding.
New approaches such as CAPEX vision AI aim to better align spending with results, though adoption remains limited. As AI systems gain independence and costs keep rising, the split is becoming clear.
The AI capex race heading into 2026 may deepen inequality, rewarding infrastructure owners while others struggle to keep up.
Inside Telecom provides you with an extensive list of content covering all aspects of the tech industry. Keep an eye on our Intelligent Tech sections to stay informed and up-to-date with our daily articles.