Full Breakdown
AI Token Strategy Drives Revenue Growth for China’s State Telecoms
8/22/2026, 2:55:10 AM
Revenue and Operational Highlights
China Mobile, China Telecom and China Unicom—all state-owned telecom operators—have placed artificial-intelligence (AI) tokens at the centre of their first-half-year financial disclosures released this month. China Telecom reported a 95 % jump in intelligent-computing revenue, lifting its broader “intelligent business” revenue 7.1 % to 31.1 billion yuan (?US$4.6 billion). Cloud revenue at the carrier rose 7.8 % to 61.8 billion yuan. China Unicom posted computing-power revenue of 41.9 billion yuan, a 13 % increase that spans data-centre services, leased compute, business software and cloud-based AI.
Platform Expansions and Token Ecosystems
All three firms highlighted token-based offerings as a key growth driver. China Telecom introduced token packages for mobile users nationwide and upgraded its Xingchen TokenHub platform, now hosting 142 large-language models and more than 420 industry-specific AI applications. The carrier also enhanced its Xirang orchestration software, extending support to over 20 chip architectures. China Unicom launched a “token supermarket” within its UniAI platform, creating an end-to-end ecosystem for creating, transferring, storing and using AI tokens. UniAI now connects more than 200 large-language models and aggregates over 500 terabytes of high-quality data.
Implications for China’s AI Market
The emphasis on billable AI tokens reflects accelerating AI adoption and soaring computing demand across China. By monetising token usage—the fundamental unit of data processed by AI models—the telecom giants are diversifying revenue streams beyond traditional voice and data services. The expansion of token-centric platforms and orchestration tools suggests a broader push to integrate AI workloads into existing network infrastructure, potentially shaping the competitive landscape for cloud and AI service providers in the country.
Outlook
The three operators intend to continue expanding token ecosystems and supporting a wider range of chip architectures, signalling ongoing investment in AI-driven services as a core component of their medium-term growth strategies.
