China’s ‘AI Plus’ Strategy Shifts to Large-Scale Industrial Deployment
China’s “AI Plus” (人工智能+) initiative is transitioning from pilot projects and experimental phases into large-scale industrial implementation, marking a pivotal new chapter in the country’s artificial intelligence industrialization efforts. The shift was formally signaled in April 2026 when the Central Politburo upgraded the initiative’s language from “deepen implementation” to “comprehensively implement,” according to Xinhua News.
Policy Escalation and Strategic Significance
The upgrade represents a decisive inflection point. “The proposal to ‘comprehensively implement the AI Plus initiative’ means that the top-level design has taken shape, the technology foundation has been basically consolidated, and conditions for industrial application are maturing,” said Zhu Keli, Founding Dean of the Guoyan New Economic Research Institute, as reported by China News Service. “It is no longer limited to pilot projects in individual fields or regions, but has risen to a normalized strategic deployment across the entire nation, all industries, and all scenarios.”
The policy journey has been rapid. First unveiled in the 2024 government work report, the initiative was codified in August 2025 when the State Council issued Guofa [2025] No. 11, titled “Opinions on Deeply Implementing the ‘AI Plus’ Initiative,” available on the State Council website. By January 2026, eight ministries had issued an “AI + Manufacturing” Special Action, and the 15th Five-Year Plan recommendations called for advancing the initiative across the board.
From Training to Inference: The Token Economy
A central theme emerging from the AI+ Ecosystem Conference (AIEC 2026), hosted by Tsinghua University’s Global Industry Research Institute in Beijing, is that AI is moving from the training phase into the inference phase. Shan Zhiguang, Director of the Information and Industrial Development Department at the National Information Center, stated that the exponential growth in Token calls serves as the most direct quantitative indicator of large-scale AI application.
“Token is no longer just a technical unit for large language models to process text,” Shan explained, as reported by Sina Tech. “It has become a new economic unit bridging electricity, computing power, model services, and application value.” This shift is driving AI infrastructure evaluation standards away from peak computing power toward unit energy efficiency, cost-effectiveness, and scenario service capability.
Market Projections and Adoption Metrics
The scale of China’s AI transformation is staggering. IDC China President Huo Jinjie projected that China’s generative AI market will approach nearly US$100 billion by 2029, with a compound annual growth rate of 68% from 2024 to 2029, making China one of the world’s most active AI application markets.
China’s AI core industry has already exceeded 1.2 trillion yuan (approximately US$165 billion), nurturing benchmark enterprises in intelligent computing, large models, high-quality dataset construction, and embodied intelligence. The country has 440 million monthly active users of AI-native apps, with daily Token calls reaching approximately 140 trillion. Notably, 93% of Chinese workplace respondents use AI tools, far exceeding the global average of 58%.
Model costs have plummeted from an average of US$60 per million tokens in early 2022 to less than US$1, laying the foundation for widespread deployment. China also accounts for approximately 60% of the world’s AI patents, ranking first globally.
The Rise of AI Agents and the ‘Humagent’ Concept
Perhaps the most transformative development is the emergence of AI agents as a new form of digital labor. Peng Zhen, Chairman of Inspur Information, argued that the AI industrial revolution has for the first time changed the definition of “labor.”
“As agents become an important part of enterprise productivity, the focus of AI-native transformation is shifting from individual efficiency improvement to organizational evolution,” Peng said at the conference. He proposed the concept of “Humagent” (Human + Agent), where enterprises evolve from managing humans to managing hybrid human-agent teams, redefining roles, permissions, and performance evaluation to maximize AI’s intellectual contribution while maintaining operational stability.
Tencent Senior Product Expert Wang Shengjie noted that the bottleneck for enterprise agent implementation has shifted from comprehension to execution capability. “What enterprises need more are digital employees who can actually work, are reliable, and controllable,” Wang said. “From understanding to execution — this is a qualitative change from ‘being able to chat’ to ‘being able to work.’”
Human-AI Complementarity
Tsinghua University’s Peng Kaiping, Dean of the Global Industry Research Institute, offered a measured perspective on the human dimension. “The ultimate goal of technological progress is to enhance human well-being, not replace human value,” Peng emphasized. “In the intelligent era, aesthetic sense, creativity, and empathy are core human traits. Agents are naturally complementary to humans — they handle standardized, deterministic execution work while humans focus on aesthetic creation, strategic decision-making, and emotional connection.”
Forward-Looking Targets
The State Council’s policy framework sets ambitious targets: by 2027, AI application penetration rates should exceed 70% across six key sectors — science and technology, industrial development, consumption upgrade, public well-being, governance capacity, and global cooperation. By 2030, penetration should exceed 90%, and by 2035, China aims to fully enter a new stage of intelligent economic and social development.
What to Watch
As China’s AI Plus strategy scales, several key questions remain: How will the Humagent organizational concept be implemented across different industries? What regulatory frameworks will govern enterprise AI agent deployment? And how will the projected 68% CAGR for China’s generative AI market be sustained amid global competition? The coming months will reveal whether China can deliver on its ambition to achieve the leap from AI technological breakthroughs to large-scale implementation, building what officials describe as a full-chain industrial competitive advantage.