The AI (artificial intelligence) race is in full swing, and the wind is blowing towards power batteries. AI technologies such as computer vision, machine learning, cloud computing, and big data are becoming key means to enhance the extreme manufacturing system. In fact, lithium-ion batteries belong to typical complex large-scale systems, exhibiting the characteristics of interdisciplinary influence. For example, the crystal structure of materials and solid-state reactions involve fundamental disciplines such as solid-state physics and electrochemistry, while charging and discharging involve physical and chemical reactions at spatial, temporal, and energy scales, and the structure-activity relationship between them is extremely complex. Although lithium batteries have emerged as valuable single point design tools in electrochemical simulation and other scenarios, there are still many practical gaps in material characterization and mechanism understanding throughout the battery life cycle. From the perspective of reasons, firstly, due to the natural constraints of several mature material systems, substantial breakthroughs in new formulas and processes are limited. Secondly, through the trial production of soft packs and button batteries for comparative testing, not only is there a large investment of manpower and material resources, but the actual effect is also often difficult to find patterns due to excessive external interference. However, with the breakthrough of the new generation of AI technology, this situation is undergoing profound changes, and generative AI has shown tremendous potential for disruptive development. Recently, South Korean battery supplier LG announced that it will use AI to design batteries for its customers, which can produce battery cells that meet customer requirements within one day; The Microsoft Quantum Computing team in the United States combined high-performance computing with AI computing and locked in a candidate material called "N2116" in just 80 hours. If traditional screening methods were used, achieving this result could take over 20 years. Domestically, CATL is also collaborating with technology suppliers such as Intel to create a battery defect detection solution that spans cloud edge end and integrates computer vision, deep learning, and machine learning technologies based on lithium battery online detection scenarios. This obviously also involves the deep application of AI technology. The author believes that the powerful computing and analytical capabilities of AI are bringing tremendous changes to battery manufacturing, helping companies control cost investment, shorten research and development cycles, and create the "best formula" for power batteries through three major processes: material selection, device design, and optimized production. More importantly, the introduction of AI will greatly promote the intelligent upgrading of key industries, empower the industrial manufacturing system at a high level, accelerate the formation of new quality productivity, and provide strong support for the construction of a manufacturing power, a network power, and a digital China. In terms of material selection, it is widely recognized in the industry that the core of technological competition for power batteries in the next decade lies in materials. In this context, most domestic lithium battery manufacturers are using computer simulation to carry out material selection, electrode and cell design, in order to help enterprises reduce the number of experiments and significantly accelerate the research and development speed of new and all solid state batteries. In terms of battery design, taking two types of new batteries as examples, one of the difficulties in developing stable electrochemical material systems for all solid state batteries is currently. Toyota in Japan has attempted to apply tens of thousands of electrolytes to batteries over the past 30 years, but has not yet successfully mass-produced them; At the same time, due to the similar radii of iron and manganese ions, atomic level mixing can be achieved. The industry is trying to add manganese elements to lithium iron phosphate to obtain better performance lithium manganese iron phosphate, but the proportional relationship between the two is difficult to overcome. The introduction of AI is expected to greatly simplify the battery design process and quickly solve the above problems through efficient simulation. In terms of optimizing production, with the development of AI algorithms and big data analysis technology, various aspects of lithium battery production can be deeply optimized. By establishing a data sharing system, analyzing multi-source databases, and optimizing production processes, manufacturers can monitor various parameters in real-time during the production process, accurately predict and adjust the production process, and effectively improve the production efficiency and product quality of lithium batteries. (New Society)