Huawei proposes a three-step path for industry intelligence, solving problems for the implementation of industrial AI.

In an era when the wave of intelligentization is sweeping across the globe, every enterprise is embracing AI. However, only a handful can truly transform it into a core competitive advantage. While many manufacturers are still competing in terms of the number of model parameters and the novelty of algorithms, a more pressing issue is how to put AI into action in real - world business scenarios and create tangible commercial value.

This is also the core proposition that Huawei is trying to address. On September 19th, during the Huawei Connect 2025, Chen Lei, the Senior Vice - President and President of Enterprise Sales at Huawei, shared Huawei's practices and insights in the process of intelligent transformation. When AI becomes the core driving force for business innovation, how should enterprises find the most suitable path for intelligent upgrading?

Therefore, Huawei has shared its practical experience and, in collaboration with partners, released nine industry solutions covering government affairs, education, healthcare, finance, manufacturing, and other sectors.

Huawei's practices are becoming a replicable and implementable industry standard paradigm, helping enterprises in various industries find the most suitable transformation paths.

Five Key Findings in Intelligent Transformation

According to the prediction of the well - known market research firm Gartner, by 2027, 80% of Chinese enterprises will deploy local multi - model AGI. By 2029, 60% of enterprises will integrate AI into their production systems, products, and services, which will become the main driving force for revenue growth.

But can enterprises simply adopt existing technologies? In reality, there is still a huge gap between technology supply and business value, which is the core challenge currently faced by enterprises in intelligent transformation.

Based on Huawei's and its partners' practical experience, Chen Lei shared five key findings in the process of intelligent transformation:

First, scenario selection is crucial. The value of AI lies more in its in - depth integration with core production scenarios, thereby reshaping processes and driving the delivery of intelligent products and services.

Second, the quality of vertical domain data determines the capabilities of industry models. Relying solely on out - of - the - box general models is insufficient for the implementation of industry intelligentization. Enterprises must train and optimize general models based on a large amount of high - quality data accumulated by themselves and their industries to form exclusive industry models and build differentiated competitiveness.

Third, AI Agents are rapidly scaling up, driving a strong demand for large - scale inference.

Fourth, human - machine collaboration is becoming a new organizational paradigm.

Fifth, systematic governance and risk management are the bottom lines that must be safeguarded. An effective governance mechanism must be established to ensure the safe, sustainable, and trustworthy application of AI.

These findings are based on Huawei's real - world insights. Take scenario selection as an example. It emphasizes that technology must serve the essence of business rather than being used for fragmented technological embellishments. This is also the strategic starting point and the rudder for all actions. The emphasis on vertical domain data also serves as a reminder to the industry: general models are like "rough - finished houses," while profound industry data accumulation is the best building material for shaping them and the foundation for the quality of intelligentization. AI Agents are the core carriers for implementing applications.

Huawei's insights also echo the current situation of the intelligent transformation wave. In fact, many seemingly common intelligent transformation needs present completely different pain points and bottlenecks in the specific business processes of each industry and enterprise. This means that there is almost no one - size - fits - all standardized solution. A solid technological foundation and in - depth industry know - how are the prerequisites for the success of intelligentization.

This also means that both technology providers and enterprise decision - makers need to take one more step forward.

During the Huawei Connect event, Chen Lei also shared Huawei's "three - step" path for promoting industry intelligentization, providing reference experience for enterprises on their transformation journey: evaluating high - value scenarios, calibrating models with vertical industry data, and deploying AI Agents on a large scale to reshape key businesses.

Make AI the Key Engine for Driving Growth

The "three - step" path is also a systematic project to transform AI from a technological concept into business value. To achieve this project, an ICT infrastructure oriented towards AI is indispensable, covering the entire process from data preparation, transmission, model training and inference, to model development. Only by building a solid intelligent foundation can complex and diverse business scenarios be supported.

This is where Huawei's strength lies. For example, in terms of data governance, Huawei can provide a complete toolchain, including the ModelArts one - stop AI development platform and the DataArts data governance and integration platform, to help enterprises transform raw data into knowledge and then knowledge into models.

Huawei's transformation practices in multiple industries such as energy, manufacturing, finance, education, healthcare, and transportation are the best examples.

For a long time, the power industry has been highly dependent on manual labor. Equipment inspection, fault diagnosis, and handling are not only inefficient but also pose significant safety risks. Based on the Ascend computing power platform and the MindSpore AI framework, Huawei and China Southern Power Grid have developed the first self - controllable large - scale model in the power industry, "MegaWatt." By integrating computer vision (CV) and natural language processing technology (NLP), it is deployed on the Ascend large - scale expert parallel cluster.

Thanks to Huawei's AI capabilities, "MegaWatt" has overcome multiple technical thresholds such as software - hardware adaptation, pre - training, and model fine - tuning. It can process 100 problem images per minute, with an efficiency 10 times that of traditional artificial intelligence algorithms. Moreover, it can automatically generate disposal plans for grid abnormalities in seconds and respond to the 15 - minute power market adjustment demand in a timely manner. Such an efficiency improvement is undoubtedly difficult to achieve with general large - scale models.

In the healthcare industry, AI has become a key factor in liberating productivity. RunDa Medical, a well - known in - vitro diagnostic service provider in China, has developed an AI medical record all - in - one machine solution based on the Ascend inference server, targeting the pain point of the heavy workload of medical record writing for doctors.

Taking West China Hospital of Sichuan University as an example, the actual deployment of this AI - generated medical record system in the hospital shows that it can generate a medical record in as fast as 1 second, support personalized configuration and optimization by doctors, with an average of no more than 4 modifications during the doctor's review stage, and support one - click writing back to the HIS system, greatly improving the diagnosis and treatment efficiency.

It is not difficult to find that in the wave of intelligent transformation, AI is no longer just a single - point innovation for icing on the cake. Instead, it is the key engine for driving business growth and even the lifeline running through the entire value chain of R & D, production, and service. Only by finding the right scenarios and consolidating the data foundation can AI play its maximum value.

Be the Fertile Soil for Industry Intelligentization

Why is Huawei the preferred choice for these enterprises?

Huawei is one of the earliest domestic manufacturers to achieve digital transformation. Huawei's AI capabilities have been widely applied and verified in its own production, logistics, R & D, marketing, and other scenarios. What has been accumulated is not only technology but also valuable industry know - how. To date, Huawei is also one of the few manufacturers in the industry that can provide end - to - end solutions from underlying computing power, operating systems, development platforms to industry applications. For example, in the fields of data storage, computing, and networking, Huawei can provide a series of integrated products, including AI storage and UCM plugins, 800GE high - speed data center network solutions, high - reliability StarLink optical modules, and the Ascend 384 super - node cluster.

This full - stack technology layout ensures in - depth collaboration and optimization among different layers of the technology stack, greatly reducing the complexity and uncertainty of system integration. It is exactly the one - stop, high - performance, stable, and reliable AI foundation that enterprises need most at present.

However, deploying AI is a complex systematic project. Coupled with the diverse and complex needs of various industries, no single enterprise in the world can handle the intelligent transformation of all industries on its own. This is where another key player comes in: partners.

If Huawei's AI platform, computing power, and framework are like a top - level scalpel, partners are like skilled surgeons. They have a better understanding of the processes, pain points, and compliance requirements of specific industries and can "translate" general AI technologies into customized solutions for solving practical problems. It can be said that partners are the last mile connecting technology and business value.

In the industry, Huawei has a rich partner ecosystem. As of now, Huawei's ecosystem has brought together more than 6,300 Kunpeng partners, more than 2,700 Ascend partners, more than 70 industry consulting firms, and 750 independent software vendors (ISVs). The continuous investment of partners tests a company's ecological appeal and cohesion, which is essentially a trust in Huawei's technological capabilities and the concept of sharing interests.

To some extent, Huawei is also one of the enterprises most proficient in building ecosystems. From the consumer market to the business market, Huawei has verified such a model countless times: attract partners with technology and brand influence. As the ecosystem grows larger and extends further, it can quickly absorb the latest needs and experiences, which in turn can feed back into its own capabilities. Then, more solid technologies can be introduced to the entire industry, forming a positive growth flywheel.

Today, this logic still works in the process of AI implementation in industries. Huawei's practices have also set a benchmark for the entire industry: the ultimate proposition of artificial intelligence has never been about whether it can be achieved, but about how to benefit people and how to co - prosper with industries and society. When the wave of intelligentization arrives, technologies will be iterated, and models will be updated. However, building a "black soil" for the industry to thrive is of far greater long - term value than delivering a single tool.

What Huawei is doing is using technology as fertilizer and an open ecosystem as soil to jointly cultivate a lush forest of intelligent innovation with thousands of partners.