Since its inception during MWC 2018, GIO has been dedicated to fostering exchanges and cooperation among industry organizations, promoting common frameworks and standards to break down silos and achieve interoperability. As AI profoundly transforms industries, the focus has shifted from digital transformation to intelligent empowerment. We should embrace AI with a positive attitude, leveraging its potential to surpass human capabilities and empower industries — driving higher efficiency, lower costs, and better products. At the same time, we must address potential risks related to accuracy and security through global collaboration. GIO will continue to serve as an open platform for sharing experiences and building consensus, helping industry organizations stay connected and aligned in a complex environment. Our goal is to support the development of unified, open standards that contribute to the growth of industries worldwide.

AI is moving from static models toward agentic AI, and the real challenges lie in trust, coordination, and control across multi-agent systems. Everyone is holding onto their own datasets and closed ecosystems, with no unified standards or safety mechanisms. To move toward commercialization, we must rely on strategic partnerships and trust-building within supply chains; otherwise, a loss of control or a simple misstep could lead to systemic failures. As we advance agentic AI, we must prioritize ecosystem integrity and robust safeguards.

Manufacturing is transitioning toward intelligence, green development, and integration. Artificial intelligence will drive the shift from automation to intelligent manufacturing, enabling a closed-loop data and decision-making process across the entire lifecycle. Current challenges include data silos, model deployability, trust and compliance, as well as organizational alignment. The future will unfold in three stages — point-based intelligence, process intelligence, and system intelligence — leading to autonomous factories and adaptive supply chains.

AI significantly reduces time and cost across all stages of ultra-high-definition content production, from pre-production to color grading, encoding, and storage. Key challenges facing the industry include the difficulty of distinguishing AI-generated content, highlighting the urgent need for a unified labeling system, as well as the massive volume of ultra-high-definition data, which continues to pose pressure on storage and metadata management.

AI is evolving from a tool into a partner, driving the Fourth Industrial Revolution across industries and reshaping finance, energy, manufacturing, healthcare, and transportation. Huawei’s “ACT” approach focuses on high-value scenarios, industry-specific models, and AI agent deployment to build enterprise-grade AI architecture. AI will accelerate human-machine collaboration and shape a fully intelligent world, calling for an open ecosystem to achieve shared success.

We have witnessed artificial intelligence evolve from isolated applications to deep industry integration driven by the flywheel effect of data, models, and applications. On the technology front, reasoning models and agent engineering are advancing rapidly; on the application front, the shift from “+AI” to “AI+” is delivering efficiency gains across entire scenarios. Key challenges remain in infrastructure adaptation, high-quality data engineering, industry-specific model migration, and agent architecture integration. CAICT and the Artificial Intelligence Industry Alliance (AIIA) will continue working with the industry to promote deep integration of AI technologies with various sectors, establish systematic pathways for intelligent transformation, and achieve agile iteration with closed-loop value creation.

From a data-centric perspective, CESI supports the data industry and sectoral applications by aligning with the policy framework of the National Data Bureau and advancing standardization for efficient use of data resources. We build data infrastructure with a focus on interoperability and unified identification, promote trusted data spaces and computing power systems, and prioritize high-quality datasets to establish methodologies and evaluation standards for “Data for AI”. We also cultivate the data factor ecosystem, facilitate data productization and circulation, and actively engage in international standardization efforts while deepening exchanges with Europe on data spaces. Through the national standardization technical committee, we systematically advance standards for data governance, technology, and sectoral applications.

AI adoption is growing rapidly, yet governance consistently lags behind. Governance should not be seen merely as a compliance obligation but as a key enabler of opportunity. By establishing a governance framework grounded in standards and risk management, organizations can scale AI innovation safely, build value chain partnerships, and enhance model performance and security. The International AI Governance Association works to promote global harmonization, reduce fragmentation across regions, and has delivered practical outputs such as the Multi-Actor Governance Framework, a layered approach to transparency, and cybersecurity governance guidance. Governance must be embedded throughout the full AI lifecycle — from design to retirement — serving as a foundational element for responsible and sustainable development.