The implications extend far beyond China’s borders. As these telecom operators position themselves as the essential intermediaries between AI developers and end-users, they’re creating a new economic paradigm where computational power becomes as fungible as currency. Analysts warn this could reshape not just China’s tech ecosystem but global cloud computing dynamics, potentially challenging Western giants like Amazon Web Services and Microsoft Azure in emerging markets.

What began as experimental pilots has now crystallized into a coordinated industry push, with each operator unveiling comprehensive token strategies at major industry conferences this year. The question now isn’t whether tokenization will succeed, but which operator will emerge as the dominant standard-setter in what could become the world’s largest AI infrastructure market.

The Tokenization Revolution: Why Telecoms Are Becoming AI Infrastructure Providers

For decades, China’s three telecom operators—China Mobile, China Telecom, and China Unicom—have competed primarily on network quality, coverage, and consumer pricing. But as traditional voice and data revenues plateau, these state-owned enterprises are making a bold bet: they will become the essential infrastructure providers for artificial intelligence.

The centerpiece of this transformation is tokenization—the process of converting computational resources, model access, and cloud services into standardized, tradable units called “tokens.” This approach mirrors cryptocurrency models but applies them to AI infrastructure, creating a new economic layer where computational power functions like a digital commodity.

“This isn’t just about selling more cloud capacity,” explains Dr. Li Wei, a senior researcher at the China Academy of Information and Communications Technology. “It’s about creating a new economic system where AI development, deployment, and consumption are all mediated through these tokenized resources. The telecoms are positioning themselves as the essential intermediaries in this new economy.”

The strategic importance lies in control. By tokenizing their vast computational resources—including the 92.5 exaFLOPS of AI processing capacity China Mobile now operates—they can monetize access to this infrastructure at scale, while also gaining visibility into AI usage patterns across industries.

From “Selling Connectivity” to “Selling AI Access”

The shift is most evident in how each operator is framing its token strategy:

• China Mobile has positioned itself as the aggregator, building what it calls a “Token operating ecosystem” that connects disparate AI models, computational resources, and service providers under a unified platform called MoMA (Mobile Model Aggregation). The company claims this approach reduces per-unit token costs by approximately 30% through economies of scale.
• China Telecom has taken a more vertically integrated approach with its Star TokenHub 1.0 platform, which it unveiled at the Digital China Summit in May. The platform treats tokens as the fundamental unit of exchange within its intelligent cloud ecosystem, allowing developers to pay for computational resources in real-time.
• China Unicom is focusing on the “Agent+Token+AI Cloud” framework, where tokens serve as both a payment mechanism and an access control system for AI services deployed across its cloud infrastructure.

What these approaches share is a fundamental reimagining of how AI services are consumed. Rather than purchasing cloud capacity by the hour or paying for specific model licenses, developers and enterprises would acquire tokens that can be spent flexibly across different services—much like purchasing credits for a subscription service.

How the Token Economy Works: A Practical Example

To understand the practical application, consider how a mid-sized Chinese enterprise might interact with this new ecosystem:

1. Resource Acquisition: The company purchases tokens from one of the telecom operators, either through direct purchase or by earning them through data contributions or other value exchanges.
2. Model Access: Using the operator’s platform (like China Mobile’s MoMA), the company selects from hundreds of pre-integrated AI models—ranging from natural language processing to computer vision—each associated with a token cost.
3. Computational Execution: The tokens are automatically converted into the required computational resources on the operator’s servers, with usage tracked in real-time.
4. Service Integration: The resulting AI outputs can then be integrated into the company’s products or services, with additional tokens potentially required for deployment or scaling.
5. Ecosystem Participation: By contributing their own AI models or data to the platform, the company can earn additional tokens, creating a feedback loop that incentivizes participation.

This model creates several advantages:

• Flexibility: Companies can mix and match services without being locked into specific providers.
• Cost Efficiency: The telecoms’ ability to aggregate resources reduces individual costs.
• Standardization: Unified token systems reduce the fragmentation that currently plagues AI development.
• Data Insights: Telecom operators gain unprecedented visibility into AI usage patterns across industries.

The Competitive Landscape: Who’s Ahead in the Token Race?

While all three operators have announced comprehensive token strategies, China Mobile appears to be taking the most aggressive approach, having already:

• Launched its MoMA platform with integration of over 300 AI models
• Achieved approximately 30% cost reductions in token operations through aggregation
• Formed industry alliances to establish standardization
• Outlined a four-phase strategy: “Aggregation, Foundation, Activation, and Shared Value Creation”

China Telecom, meanwhile, has positioned itself as the innovator in token technology, with its Star TokenHub platform designed to be the operating system for the token economy. The company has emphasized international expansion, suggesting it may seek to export this model to global markets where it has existing infrastructure.

China Unicom, while slightly later to the party, is leveraging its strengths in enterprise services to position its token framework as particularly attractive to large corporations and government agencies.

Industry Impact: Who Wins and Who Loses?

The telecoms’ tokenization strategies have significant implications across the tech ecosystem:

Winners

• AI Developers: Lower costs and standardized access to computational resources could accelerate innovation.
• Enterprise Customers: Flexible token-based pricing models may reduce long-term AI infrastructure costs.
• Telecom Operators: New revenue streams from token sales could offset declining traditional telecom revenues.
• Chinese Tech Ecosystem: Standardization could reduce fragmentation and encourage domestic AI development.

Potential Challenges

• Regulatory Uncertainty: While tokenization avoids cryptocurrency classifications, regulators may still scrutinize these systems.
• Interoperability: Competing standards from different operators could create new fragmentation.
• Competition with Cloud Giants: Western providers like AWS and Azure may respond with their own tokenized offerings.
• Consumer Adoption: Enterprises may be more willing to adopt token systems than individual consumers.

Analysts at Analysys International warn that while the concept is compelling, successful implementation will require overcoming significant technical and market challenges. “The telecoms have the infrastructure, but they lack the software ecosystem and developer community that cloud providers like Alibaba Cloud and Tencent Cloud have built,” notes a recent report.

The Global Implications: Could This Model Spread?

While the immediate focus is on China’s domestic market—where these operators already control 99% of fixed-line and mobile networks—the tokenization model has potential global applications. Several factors could determine its international reach:

• Regulatory Environment: Countries with similar state-backed telecom structures (e.g., Russia, India, Southeast Asia) may be early adopters.
• AI Demand: Emerging markets with growing AI adoption could find token models particularly attractive.
• Competitive Response: Western cloud providers may develop competing tokenized services.
• Standardization: If China’s operators can establish industry-wide standards, they may export this model as a package.

Some observers suggest this could create a new paradigm for global cloud computing, where regional operators become the essential infrastructure providers rather than Western hyperscalers dominating all markets. “We’re seeing the emergence of a multipolar cloud computing landscape,” says Margaret O’Malley, a senior fellow at the World Economic Forum. “The telecom-led token model could gain traction in regions where data localization and sovereignty are priorities.”

What’s Next: The Road Ahead for Tokenization

The next 12–18 months will be critical in determining whether tokenization becomes the dominant model for AI infrastructure. Key developments to watch include:

• Industry Alliances: Will the three operators coordinate their token systems, or will competition lead to fragmentation?
• Pilot Programs: Which enterprises and government agencies will be the first to adopt these systems at scale?
• Regulatory Clarity: How will authorities classify and regulate these token systems?
• Technical Development: Will the platforms achieve the promised cost efficiencies and performance improvements?
• Global Expansion: Will any operator attempt to export this model to international markets?

The telecom operators themselves have outlined ambitious timelines. China Mobile, for example, has indicated it plans to expand its MoMA platform to include thousands of additional models within the next two years, while China Telecom has suggested it will seek international partnerships to promote its Star TokenHub globally.

Conclusion: A New Era for China’s Telecom Industry

The tokenization strategies unveiled by China’s three telecom giants represent more than just a business pivot—they signal the beginning of a new economic era where computational resources become as tradable as traditional commodities. By converting their vast infrastructure into a tokenized ecosystem, these state-owned enterprises are positioning themselves not just as connectivity providers but as the essential infrastructure layer for China’s AI future.

Success in this endeavor could redefine the global cloud computing landscape, creating a multipolar system where regional operators play a more central role. For now, the focus remains on domestic implementation, with each operator racing to establish its token platform as the industry standard.

The next major checkpoint will be the World Mobile Congress 2026 in Barcelona (scheduled for October 2026), where observers expect all three operators to provide updates on their tokenization progress and potentially announce new partnerships. This event will be crucial in assessing whether tokenization is progressing toward becoming the dominant model for AI infrastructure—or if significant challenges remain to be overcome.

As China continues its technological ascendance, one thing is clear: the telecom operators that once competed primarily on network speed are now locked in a high-stakes race to control the very foundation of artificial intelligence. The winner in this competition won’t just be a telecom leader—they’ll be shaping the future of global computing itself.