—Breaking New Ground to Industrial Intelligence
SHANGHAI, Sept. 20, 2025 /PRNewswire/ — Leo Chen, Huawei’s Senior Vice President and President of Enterprise Sales, shared Huawei’s latest insights and practices for industrial intelligence today at HUAWEI CONNECT 2025, including the company’s three-step “ACT” pathway for intelligent transformation. During Chen’s keynote, titled “Breaking New Ground to Industrial Intelligence”, Huawei also launched nine major solutions for industrial intelligence that it developed alongside its partners.
Five key findings driving industrial intelligence
Leo Chen, Senior Vice President, President of Enterprise Sales, Huawei
AI technology has developed rapidly over the past few years. Chen noted that this has raised three major questions for enterprises:
- How can enterprises ensure their AI investments generate real business value?
- How can enterprises use their own data to defend their competitive advantages?
- How can enterprises help AI use cases move beyond pilots to large-scale application?
Chen shared how Huawei is tackling these questions with three success stories.
First, Huawei developed a dynamic master–sub-agent architecture for the banking industry that incorporates systematic engineering capabilities into banking workflows. Huawei helped their customer accelerate AI application using this architecture in a number of scenarios like interactions, transactions, and risk control. This has enabled the customer to increase revenue and efficiency, while more effectively managing risk.
Second, China Southern Power Grid used Huawei’s Ascend computing platform and MindSpore AI framework to develop a large model that it named “MegaWatt”. This model runs on an Ascend MoE Expert Parallelism cluster and combines computer vision and natural language processing. It has helped the customer improve defect and risk identification efficiency during intelligent power line inspections by five times, and increased image recognition accuracy to over 90%.
Third, Runda Medical used Huawei’s Ascend inference servers to develop an AI medical record appliance solution that improves both the quality and efficiency of medical recordkeeping. Since its deployment at West China Hospital, the solution has helped medical record generation time to about one second, which has dramatically improved the hospital’s consultation efficiency.
Chen then shared five key findings Huawei developed over the course of these and other intelligent transformation projects.
First, choosing the right scenarios for implementation is critical to successful transformation. The value of AI lies in its ability to transform processes and promote intelligent product and service delivery when deep integrated with core production scenarios.
Second, the capabilities of industry-specific models depend on high-quality vertical data. General-purpose models often struggle when applied in industrial scenarios. The most successful models are those that are trained and fine-tuned based on massive amounts of high-quality proprietary data —which transforms them into industry-specific models. Such industry-specific models become a significant competive edge for enterprises.
Third, the demand for large-scale inference will continue to grow fast alongside AI agents, fueling demand for large-scale inference.
Fourth, human-AI collaboration is becoming a new organizational paradigm.
Fifth, systematic governance and risk management are becoming more critical than ever, driving many organizations to pursue more effective governance. The goal of this governance is to ensure secure, sustainable, and trustworthy AI.
The “ACT” pathway for industrial intelligence
Huawei’s proposed three-step “ACT” pathway is designed to promote AI adoption across industries. ACT stands for: Assess high-value scenarios, Calibrate AI models using vertical data, and Transform business operations with scaled AI agents.
Huawei assesses high-value scenarios using its AI Scenario Assessment Framework to evaluate business value, scenario maturity, and business–technology integration. This framework has already helped customers identify more than 1,000 core production scenarios for AI adoption.
The company also provides a full toolchain and AI security protection system to help enterprises calibrate general models and create industry-specific models. This toolchain uses vertical data to transform raw enterprise data into knowledge, and knowledge into models.
For the final step of the ACT pathway, transforming business operations with scaled AI agents, Huawei has a one-stop Versatile platform that can automatically generate agents, as well as workflows with more than 100 steps. This platform accelerates agent deployment. The company has additionally developed an AI talent enablement program that helps business professionals effectively develop, deploy, and operate AI agents.
Enterprises looking to follow the ACT pathway will need AI-oriented ICT infrastructure that covers the entire process, from data preparation and movement, to model training, inference, and development. Chen emphasized that Huawei is continuing to innovate in data storage, computing, and networks for this purpose, and is working on creating integrated products that make the process easier for customers. Examples of these products include AI storage and Unified Cache Manager plugins, 800GE high-speed data center network solutions, the company’s highly reliable StarryLink optical modules, and the Ascend AI SuperPoD 384.
Launch event: Nine jointly-developed solutions for industrial intelligence
Huawei is promoting a “Huawei + Partners” collaboration ecosystem that they are developing through three main initiatives: the opening up of its software, hardware, and systems; enablement platforms and tools for its partners; and the rapid replication of successful practices through industry experience sharing. To date, their ecosystem has attracted over 6,300 Kunpeng partners, 2,700 Ascend partners, 70 consulting firms, and 750 ISVs.
Before the keynote session concluded, Huawei unveiled nine major solutions for industrial intelligence that it developed in collaboration with partners: City AI Center & Foundation Model Solution, Intelligent Computing Labs Solution, Medical Technology Digital and Intelligence 2.0 Solution, Banking AI and Foundation Model Solution, Intelligent Manufacturing R&D Solution, SMART Logistics & Warehousing Solution, Intelligent Distribution Solution, Intelligent Exploration and Development Solution for Oil and Gas, and Steel Blast Furnace Temperature Prediction Solution.
