China Galaxy Securities: Token anchors industrial value, industrial ecology expected to be reshaped.

China Galaxy Securities released a research report stating that tokens have become the value anchor of the intelligent era, widely used as the settlement unit in the AI industry. From the demand side, the global call volume of tokens is rapidly increasing, driving the AI industry from model iteration towards commercialization and implementation, which is expected to reconstruct the production elements, industry logic, and business models of the media industry from the bottom up. Token economy empowers the traditional content production industry, with the support of AIGC tools, moving towards unlimited supply and potentially achieving prosperity in the content industry. It is recommended to focus on the core beneficiaries of the AI industry, with a focus on: 1) internet giants with comprehensive AI layouts; 2) companies leading in AI video tool products; 3) leading companies in niche fields benefiting from AI applications; 4) large model manufacturers continuously advancing technological iterations. The main viewpoints of China Galaxy Securities are as follows: Tokens are the value anchor of the intelligent era In March 2026, the Chinese Development High-Level Forum officially designated the Chinese official name of Tokens as "" (Ciyuan), positioning it as the value anchor of the intelligent era and the unified settlement unit at both ends of the industrial supply and demand. Currently, the daily call volume of tokens displays an exponential growth trend in China, signaling that the domestic artificial intelligence industry has moved beyond the model technology verification stage, entering a period of mass consumption of computing power and commercial operation. AIGC drives industrialization of content, reshaping the industry landscape with infinite supply As the consumption volume of tokens continues to rise rapidly, AI reasoning demand is fully unleashed, leading the media content industry to accelerate the fundamental transition from traditional manual creation mode to AI-native generation mode. Content production completely breaks free from constraints on human productivity, bottlenecks in creative efficiency, and limitations of high marginal costs, officially entering a stage of industrialized mass production that is scalable, standardized, and infinitely replicable. Continuous technological iteration of large models, open-source ecology promotes ecological development The continuous increase in token call volume directly drives the explosive demand for API interfaces on various application ends downstream, leading this demand to further cascade upstream to the model foundation manufacturers. The continuous increase in token demand not only amplifies the reuse value of the model base but also effectively dilutes the unit cost of model training and technical operations, continuously strengthening the technological barriers and market competitive advantages of model foundation manufacturers, becoming the core DRIVE supporting the long-term growth of the model foundation industry. Cloud service volume-price resonance, expected to open up new growth space With the continuous increase in token consumption, corresponding to the massive increase in AI reasoning demand across society, this constitutes a certain positive for cloud service providers led by Alibaba Cloud. With tokens becoming the unified value settlement unit of the intelligent era, the business model of cloud computing is undergoing a fundamental transformation, gradually moving away from the traditional IaaS model of server leasing and time-based billing to a MaaS intelligent service pay-per-use system centered around token consumption. This drives high-margin AI service revenue expansion for cloud providers, optimizing the overall profit structure, entering a upward cycle of demand expansion and volume-price resonance in the industry. Risk factors Risks related to policy and regulatory environments, risks of AI technology development falling short of expectations, risks of AI applications not meeting expectations, risks of short-term adjustments in market sentiment and fund flows.