According to reports, research firm Exponential View states that the AI industry has reached a critical inflection point, with tens of billions of dollars invested in AI infrastructure by tech companies in recent years beginning to show initial commercial validation. Data indicates that AI-related revenue from global hyperscale cloud providers and emerging cloud services is approximately $25 billion, exceeding the estimated depreciation costs associated with AI data centers and chip investments—around $21 billion—for the second consecutive quarter.

This signifies that AI industry revenue is now starting to offset the cost pressures stemming from infrastructure capital expenditures, marking a shift from capital expenditure-driven expansion toward a phase of revenue validation. Current revenue streams primarily come from AI cloud services, GPU compute leasing, large model APIs, enterprise AI software, and generative AI applications. The report also notes that due to persistently high costs in GPUs, data centers, power consumption, and model R&D, the industry remains far from a high-profit phase; the current stage is more about validating investment sustainability rather than achieving large-scale profitability.