On August 1, 2026, the National Energy Administration officially released the "Guiding Opinions on Promoting Deep Integration of Artificial Intelligence and Energy Industries" (hereinafter referred to as the "Opinions"),明确提出 that by 2030, the penetration rate of AI in the energy sector will exceed 30%, and cultivate a group of internationally competitive AI+energy leading enterprises. This policy marks the official entry of China's energy industry into an accelerated period of intelligent transformation, and also indicates that the integration of AI and energy will create a new trillion-level market.
I. Policy Background: Energy Transformation Encounters Bottlenecks, AI Becomes the Key to Breakthrough
In recent years, China's energy industry has faced dual challenges: on the one hand, under the dual carbon goals, the installed capacity of new energy has grown explosively, but the volatility and intermittency of wind and solar power have brought huge pressure to the stable operation of the power grid; on the other hand, the dispatching efficiency, operation and maintenance costs, and safety hazards of traditional energy systems are increasingly prominent. Data from the National Energy Administration shows that in 2025, the proportion of new energy power generation has exceeded 25%, but the wind and solar curtailment rate still remains at around 5%, and the utilization rate of energy storage is less than 60%. In this context, AI, as the core technology of intelligent perception, prediction and decision-making, is regarded as the key to solving the pain points of the energy industry.
The release of the "Opinions" is not accidental. As early as 2024, the National Energy Administration launched pilot demonstration projects for AI+energy; in 2025, the State Council's "New Generation Artificial Intelligence Development Plan" again listed energy as a key application area. Now, the release of this "Opinions" marks the transformation of AI and energy integration from local pilots to comprehensive expansion, with policy strength and system beyond market expectations.
II. Key Directions: Four AI+Energy Scenarios Accelerating Implementation
The "Opinions" clarifies six key tasks, among which four directions are most concerned and have the most investment potential.
1. Smart Grid: AI Transforms the Grid from "Passive Response" to "Active Prediction"
State Grid and China Southern Power Grid are accelerating the deployment of AI dispatching systems. Through deep learning algorithms, AI can perform minute-level forecasting of the entire network load with accuracy exceeding 98%; at the same time, AI can automatically generate optimal dispatching strategies and issue early warnings before failures occur. It is estimated that in the first half of 2026, AI-assisted dispatching by State Grid has reduced power outage time by about 1.2 million hours, with direct economic benefits exceeding 3 billion yuan.
2. New Energy Power Prediction: Solving the "Dependence on Weather" Problem
AI combined with meteorological big data and historical power generation data can achieve ultra-short-term prediction of output power for wind farms and photovoltaic power plants, with error rates reduced from 15% of traditional methods to less than 5%. This significantly reduces the grid connection threshold and provides a precise basis for power trading decisions. Currently, leading equipment manufacturers such as Goldwind and Sungrow have integrated AI prediction modules as standard and are beginning to transform into third-party independent software service providers.
3. Smart Energy Storage: AI Achieves the Optimal Solution for "Peak Shaving and Valley Filling"
Energy storage is the core of the new power system, but the complexity of charge-discharge strategies limits its utilization rate. AI reinforcement learning models can optimize energy storage charge-discharge strategies in real-time, considering multiple factors such as electricity prices, load, and battery health, increasing the internal rate of return of energy storage projects by 3-5 percentage points. Energy storage integrators such as CATL and BYD have all launched AI smart energy storage systems, claiming that their equilibrium degree has improved by more than 40%.
4. Oil and Gas Exploration and Safety Production: AI Replaces High-Risk Positions
In enterprises such as PetroChina and Sinopec, AI is widely used in seismic data analysis, reservoir simulation, pipeline inspection, and dangerous operation monitoring. PetroChina's AI exploration large model "Kunlun" has successfully discovered two billion-ton oil and gas fields in the Tarim Basin, shortening the exploration cycle by 30% and reducing costs by 22%.
III. Industrial Ecology: AI Giants and Energy State-Owned Enterprises Compete and Cooperate
The release of the "Opinions" has triggered a "positioning war" around AI+energy. Huawei, Baidu, iFlytek and other tech giants are entering with AI large model capabilities, forming strategic alliances with energy state-owned enterprises; while State Grid, China Southern Power Grid, PetroChina and other units rely on massive data and application scenarios to form internal AI teams, and even incubate independent technology companies.
A typical trend is that energy enterprises are transforming from "clients" to "joint creators." In June 2026, State Grid and Huawei jointly released the "Grid Large Model," with parameters reaching the tens of billions level, capable of sensing the real-time status of hundreds of millions of terminal devices simultaneously, and has been trial-run in some provincial grids. Meanwhile, Baidu Smart Cloud and PetroChina's cooperation in creating a "refining and chemical intelligent optimization system" has increased the yield of refined oil by 0.8 percentage points, with annual profit exceeding 1 billion yuan.
This competitive and cooperative relationship is reshaping the industrial chain structure. For software and algorithm suppliers, the energy market space is broad but barriers are high, requiring deep understanding of energy business; for energy equipment enterprises, AI capabilities have become a new point of differentiated competition. It can be predicted that the next two years will see the emergence of a number of professional AI+energy solution providers, and unicorn companies are expected to emerge.
IV. Investment Opportunities: Three Main Lines to Layout the Trillion-Dollar Track
From the perspective of the capital market, the imagination space of AI+energy concepts is no less than that of photovoltaics and lithium batteries in the past. According to CCID Consulting forecasts, by 2030, the scale of China's AI+energy market will exceed 2.5 trillion yuan, with a compound annual growth rate of 34%. We recommend paying attention to the following three investment main lines:
- Grid AI Intelligence: Benefiting from the digital transformation of the power grid, pay attention to the AI upgrade business of power equipment leaders such as NARI Technology, XJ Electric, and Sifang Electric, as well as power software service providers such as Henghua Technology and Jingwei Technology.
- New Energy + AI Services: In the field of power prediction and smart wind farms, pay attention to companies that have already deployed AI maintenance, such as Rongzhirixin and Riyue Co., Ltd.; for AI in energy storage, pay attention to integrators such as CATL, Sungrow, and Kehua Data.
- AI Technology Empowerment Platform: Focus on platform companies that output AI capabilities to the energy industry, such as Huawei Ascend industry chain, iFlytek, and SenseTime, which will receive order valuation premiums from energy state-owned enterprises.
In addition, be wary of short-term speculation risks. Some concept stocks have already discounted expectations, and real performance realization takes time. Investors should focus on companies with actual orders, technical barriers, and deep cooperation with state-owned enterprises.
V. Future Outlook: AI+Energy Will Give Birth to a New Energy Internet Form
Looking to the future, AI will not only be an "auxiliary tool" for the energy system, but will reshape the entire chain model of energy production, transmission, and consumption. With the integration of AI with IoT, digital twin, blockchain and other technologies, we are expected to see a highly autonomous "Energy Internet": every building is a micro power plant, every battery participates in grid interaction, AI automatically matches supply and demand, and achieves dynamic pricing and intelligent delivery of energy.
This vision requires the long-term joint efforts of policy, technology, and capital. But from the perspective of 2026, the direction is clear and the pace is accelerating. As the "Opinions" says, "AI+energy is an important starting point to grasp the new round of scientific and technological revolution and industrial change." For investors, now is a good time to make forward-looking layouts, but more patience and in-depth research are needed.
Huagang Zhishi will continue to pay attention to the policy dynamics, technological breakthroughs, and market changes of AI+energy, bringing you timely, objective, and in-depth trend outlooks. We hope this article can help you see the whole picture of this historic opportunity and gain an advantage in investment decisions.
