AI Economy Surges: $1.1 Trillion in Sales, Exceeding $1.75 Trillion Run Rate

The burgeoning generative AI economy has reached a monumental milestone, generating an astounding $1.1 trillion in sales over the past twelve months. This figure, detailed in the inaugural "State of AI Economy" report by 36Kr Global, represents a growth rate that is nothing short of staggering. On an annualized basis, the sector's revenue run rate has now surpassed an impressive $1.75 trillion, signaling a powerful and sustained expansion that dwarfs previous technological revolutions.

Unprecedented Market Valuation and Methodology

This comprehensive report marks a significant achievement as it presents what is believed to be the industry's first bottom-up, deduplicated measurement of full-stack consumer and enterprise AI spending. The research team dedicated months to meticulously building this dataset, aiming to answer critical questions about the AI wave: the market's exact size, the sustainability of its revenue growth, its ability to cover upfront investment costs, and the future economic efficiency as token prices fall and quality improves. The methodology employed to calculate the demand side is particularly noteworthy, focusing on actual dollars paid by end customers to avoid double-counting value flowing through the supply chain.

Navigating the Demand Side Complexity

While the supply side of the AI market—encompassing providers of chips, memory, transformers, and cooling systems—is relatively transparent due to the public nature of many key players, mapping the demand side presents a far greater challenge. A substantial portion of AI revenue flows to privately held companies like OpenAI, Anthropic, and Cursor, which are not obligated to disclose financial details. Furthermore, even large public cloud providers such as Amazon, Google, and Microsoft do not systematically break out independent revenue figures specifically from their AI business segments. To overcome this opacity, the researchers developed a proprietary AI economy model, carefully analyzing public statements, leaks, and corporate disclosures to derive high-confidence factual details and construct detailed financial models for key contributing companies and business units.

Key Findings: Revenue and Infrastructure Costs

The report confirms that the AI ecosystem has generated $1.1 trillion in revenue after eliminating double-counting, with a healthy growth momentum. This revenue is expanding at roughly three times the rate of the mobile internet or early internet waves. A critical question addressed is whether this AI-generated revenue can adequately cover the massive capital expenditures required for infrastructure. The research separates AI-specific capital expenditures for major hyperscalers and specialized AI cloud providers. It reveals that AI revenue directly attributable to hyperscalers currently "just barely covers" the depreciation expense for AI infrastructure, which is depreciated over 6 years for compute assets and 14 years for other infrastructure. This extended asset lifespan is justified by sustained market demand far exceeding compute capacity and the increasing sophistication of cloud providers in managing large-scale GPU clusters.

The Future of Token Economics and Quality

Looking ahead, the report delves into the evolving economics of AI, particularly as token prices decline. Demand elasticity analysis suggests that falling prices stimulate total spending, with an estimated 10% drop in price leading to a 12% to 18% increase in token usage, thereby driving overall spending upward. The researchers emphasize that while tokens are a practical billing metric, they don't fully capture the "intelligence" circulating in the industry. To provide a more objective measure, the concept of "quality-adjusted output tokens" is introduced, which accounts for token production, effective output presented to end-users, and the comprehensive capabilities of underlying models, offering a more accurate "intelligence quotient" for evaluating the AI economy.

Exclusions and Future Scope

It is important to note what the report did not include in its calculations. "Internal AI efficiency gains," such as Meta or Google increasing ad revenue through optimized recommendation systems, were modeled but excluded from this specific report. Similarly, efficiency improvements and cost savings achieved by large tech companies using internal AI tools were not accounted for. Professional services and system integration fees were also excluded, as these represent indirect investments rather than direct AI company revenue. While a model for the Chinese market was built, the initial version 1.0 report does not yet incorporate data from this region. The report also touches upon the impact of AI demand on the U.S. power industry, trends in token costs, and forecasts four potential growth scenarios for AI demand.

Broader Implications and Executive Sentiment

The findings underscore a significant shift in enterprise strategy, with executives across industries in Europe and the U.S. planning to increase their AI investments. The increasing frequency with which AI is discussed on earnings calls highlights its growing strategic importance. Notably, half of the surveyed CEOs believe their job security is contingent on delivering strong AI performance, underscoring the high stakes involved in AI adoption and implementation. This widespread executive focus, coupled with the massive economic figures, paints a picture of an AI-driven transformation that is rapidly reshaping the global business landscape.