In a recent online event held on September 17, 2026, Huawei launched its much-anticipated series of reports focusing on the future of intelligent technology, titled "Act Today. Shape Tomorrow." This event marked the debut of two significant reports: "Intelligent World 2035: Turning Vision into Action" and the "Global Digitalization and Intelligence Index 2026." The former outlines ten critical strategic axes for constructing intelligent systems designed to navigate the complexities of the impending agentic AI era.
The report emphasizes the shift of AI from the digital domain into physical spaces, making it integral to operational workflows and societal structures. This transformation is poised to redefine industry standards and facilitate widespread changes across various sectors. David Wang, Vice Chairman and Rotating Chairman of Huawei, articulated in the report the importance of agentic AI as a key element in this unfolding transformation. He stated, "For AI to reshape operational models and competitive foundations, we must confront several significant challenges. The pathway to realizing our vision is clear, but it necessitates prompt action. This theme resonates throughout the current year's report."
The report systematically delves into the profound challenges presented by the ongoing AI revolution and proposes concrete recommendations for the establishment of infrastructures necessary for an agentic world. Notably, it addresses a crucial transition from application-centered digital worlds to agent-centric environments. In these settings, agents are continuously aware of their surroundings, process information autonomously, and interact dynamically with humans in real-time. This shift anticipates a staggering projection of token consumption that could surge by a factor of 100,000 by 2035, with agent-driven traffic constituting over 90% of this figure. Such advancements will impose significant demands on computational, networking, and cloud infrastructure—far exceeding those of conventional internet applications.
To overcome the anticipated challenges, the report delineates ten priority axes:
1. Advancing to General Artificial Intelligence (GAI) equipped with autonomous cognition and versatile task capabilities as a core pathway toward achieving GAI, necessitating breakthroughs in linguistic intelligence, embodied intelligence, and scientific reasoning.
2. Enhancing computing cluster capabilities by a factor of 100 and dramatically reducing task costs for agents by 1,000, encouraging the adoption of SuperPoD computing systems to optimize token production efficiency.
3. Developing advanced storage and memory systems that ensure causal precision and reliable traceability to facilitate the transition from mere data to knowledge and memory.
4. Building intelligent connectivity infrastructures that prioritize user experience through the integration of AI within communication systems.
5. Investigating scaling methodologies and chip design tools to simultaneously enhance latency and energy efficiency, moving away from geometric scaling to time scaling for semiconductor evolution.
6. Establishing an agent operating system aimed at autonomous evolution, laying the technical groundwork for agents with capabilities summarized by (Coordination × Execution × Memory × Connectivity) ^ Evolution.
7. Defining a new paradigm of intelligence for devices that is both adaptive and ubiquitous, enabling omnipresent intelligent services across devices and scenarios.
8. Promoting optimal self-learning and security in intelligent driving technologies to advance vehicles into embodied mobile agents.
9. Extending energy and thermal limits to enhance computing power in AI-dedicated data centers.
10. Ensuring the safety of the intelligent world by addressing security and privacy concerns surrounding autonomous agents.
The report underscores that successfully meeting these challenges in the age of agentic AI requires the development of an end-to-end intelligent infrastructure focused on tokens—covering generation, distribution, consumption, and security aspects. This ambition calls for the implementation of SuperPoD systems and open ecosystems aimed at optimizing token production and establishing intelligent connectivity infrastructures to ensure seamless token flow. Furthermore, deploying agentic cloud platforms will accelerate AI adoption across all sectors, evolving tokens into tangible intelligent services that reshape user experiences and business operational models.
In conclusion, Huawei recommends five initiatives from the report, advocating for a long-term vision, human-centric focus, technology for social good, commitment to openness and collaboration, and a systemic approach to technology-driven progress in the sector. Additionally, the "Global Digitalization and Intelligence Index 2026," co-developed by Huawei and Tsinghua University's Economic Research Institute, aims to guide countries toward becoming intelligent economies. This report, fused with insights from experts in economics, sociology, and ICT, anticipates AI's financial impact, projecting to generate over $27 trillion in cumulative economic value in the next five years. The global investments in digital and intelligent infrastructures are expected to witness a compound annual growth rate (CAGR) of 19.14%, exceeding $4 trillion by 2030, establishing a robust foundation for the continued evolution of intelligent economies. The index assesses the smart and digital developments of 90 countries, categorizing their progress into creators, users, and pioneers, thereby offering policymakers clear pathways to an intelligent economy tailored to their unique contexts.
As stated in the report's preface by Wang, "Visions are articulated through words but assessed through actions. Huawei is ready to collaborate with global clients and partners to turn every meaningful measure we take today into a significant step toward an intelligent world by 2035."
For more detailed insights, visit
Huawei's Intelligent World.