China’s daily AI token calls reach 140 trillion, up more than 1,000-fold since early 2024
China's daily AI token calls surged to 140 trillion by March 2026, marking a more than 1,000-fold increase from early 2024 figures, according to the China Academy of Information and Communications Technology.
Intelligence analysis by Gemini 2.5 Flash
New data reveals an unprecedented surge in AI processing in China, with daily token calls reaching 140 trillion by March 2026. This exponential growth, attributed largely to the widespread adoption of AI agents, highlights the escalating demand for AI infrastructure and sophisticated token management systems within the country.
Imagine you have a super-smart robot helper that understands what you want and then does many tiny steps to get it done. Each tiny step costs a 'token,' like a small coin. China's robots are now doing so many tiny steps each day that they're spending 140 trillion 'coins,' which is a thousand times more than they spent just a little while ago! This is because more people are using these smart robots, and the robots are getting better at doing lots of things on their own.
Analysis
Explosive Growth in AI Processing
China has witnessed an extraordinary surge in its daily AI token calls, reaching an astounding 140 trillion by March 2026. This figure represents a more than 1,000-fold increase from the approximately 100 billion daily calls recorded in early 2024, according to data cited by Wei Liang, deputy head of the China Academy of Information and Communications Technology (CAICT). The statistics, revealed in a preview for a CCTV Finance program, underscore the immense scale at which AI models are being utilized and processed within the country.
Tokens serve as a fundamental metric for quantifying the amount of information processed by artificial intelligence models. They are also crucial for calculating the operational costs associated with running these sophisticated AI systems. The sheer volume of tokens processed daily reflects not only the growing computational demands but also the expanding integration of AI into various sectors and applications across China.
The Rise of AI Agents
The primary driver behind this monumental increase in AI token calls is linked directly to the wider adoption of AI agents. These advanced AI systems are designed to perform complex tasks autonomously, often by breaking down a single user instruction into multiple, sequential model calls. Each of these sub-calls consumes tokens, leading to a multiplicative effect on overall token usage.
The proliferation of AI agents signifies a shift towards more sophisticated and autonomous AI applications, moving beyond simple query-response systems. This trend creates a significant demand for robust token pricing and scheduling systems, essential for managing the computational resources and costs associated with these multi-step AI operations. As agents become more prevalent, the infrastructure supporting them must scale accordingly.
Implications for China's AI Ecosystem
This exponential growth in AI token calls has profound implications for China's burgeoning AI ecosystem. It suggests a rapid maturation of AI applications and a deepening reliance on AI technologies across industries. The demand for processing power, data centers, and advanced AI chips is likely to intensify, potentially spurring further domestic innovation and investment in these critical areas.
Furthermore, the need for efficient token management and cost optimization will become paramount for businesses and developers leveraging AI. This could lead to the development of more sophisticated AI governance frameworks and resource allocation strategies. The data positions China at the forefront of global AI utilization, showcasing its commitment to integrating AI into its economic and technological fabric at an unprecedented pace.
Key points
- China's daily AI token calls reached 140 trillion by March 2026.
- This represents a more than 1,000-fold increase from approximately 100 billion daily calls in early 2024.
- The figures were cited by Wei Liang, deputy head of the China Academy of Information and Communications Technology (CAICT).
- Tokens measure the amount of information processed by AI models and are used to calculate usage costs.
- The surge is attributed to the wider adoption of AI agents, which trigger multiple model calls from a single user instruction.
