AI is the battlefield, and the issue is "profit"… The worries of a big tech company that invested $600 billion

The market atmosphere surrounding AI technology is undergoing a clear shift. In the past, simply mentioning artificial intelligence could elicit enthusiastic responses from investors, but now the focus has shifted to whether its growth can directly translate into profitability. AI remains at the core of the market, but if its economic viability cannot be demonstrated, companies’ positions will face significant instability.

In recent quarters, the market has moved away from the simple equation of “AI = growth.” Investors are now more concerned with cost-effectiveness. By 2026, annual capital expenditure related to AI is expected to surpass $600 billion. Under this structure, investors are analyzing more precisely whether AI technology can be converted into actual returns, rather than just focusing on the technology itself. This change not only affects publicly listed companies but also impacts the entire tech ecosystem, from startups to non-listed companies preparing for mergers and acquisitions.

The so-called “Big Five” cloud market leaders—Apple (AAPL), Microsoft (MSFT), Amazon (AMZN), Google (GOOGL), and Meta (META)—are investing heavily in AI infrastructure. By 2026, their total capital expenditure is projected to reach $600 billion, a growth of over 36% from the previous year. About 75% of this is concentrated in AI-related infrastructure, much of which is financed through debt, a point that warrants particular attention.

Concerns about whether this over-investment can be converted into real returns are growing. Take Microsoft as an example: recent quarterly capital expenditure increased by approximately 67% year-over-year, surpassing $37 billion in a single quarter, but the growth momentum of Azure has slowed. This has led to a 21% decline in its stock price over the past six months, erasing hundreds of billions of dollars in market value.

Oracle (ORCL), despite facing rising demand for AI cloud services, announced plans to increase its capital expenditure next year to over $50 billion, sparking worries about overheating investments. With high debt levels already in place, most of its new funding is planned to be raised through debt and equity issuance.

Even NVIDIA (NVDA), at the heart of the AI era, and OpenAI have not escaped the impact of cooling investment enthusiasm. The early buzz around billion-dollar infrastructure plans has recently become uncertain, with NVIDIA stating that “nothing has been finalized yet.” To diversify supply chain risks, OpenAI is expanding partnerships with companies like AMD and Cerebras Systems.

As market vigilance against “over-investment in AI” increases, founders of AI startups also need to adjust their strategies. First, their products should not trigger additional capital expenditures but should instead focus on operating existing infrastructure more efficiently or improving profitability. Key metrics now include unit cost reduction, increased utilization of computing resources, faster deployment, and higher revenue efficiency.

Technological flexibility is also crucial. Structures that can be compatible with multiple cloud environments and models, and do not rely on specific chips or infrastructure platforms, are gaining favor in M&A markets. Finally, founders need to operate their businesses with a “public company mindset” from now on. Only with a sound profit model, verifiable key indicators, and a sustainable long-term strategy can they survive after the AI boom subsides.

Today, AI is no longer a choice but a battlefield. To survive in this war, precise strategic planning, financial management skills, and especially clear answers regarding profitability are indispensable. Since the rules of the AI era have changed, companies can no longer follow the formulas of the past.