From the perspective of healthy industrial development, cutting-edge technological innovation requires extensive international cooperation, and open decoupling can form a more prosperous industrial and commercial ecology.

Cui Li believes that any technological innovation breakthrough, commercial closed loop is the foundation of its vitality, the final customer, whether C-end or B-end, will be placed in their own landing scene,

goals and resource constraints under the balance consideration, the core is to solve the problem to create value, suitable is the best, so "training and pushing" and "virtual and real" is the only way to promote industrial upgrading AI.

Talking about how artificial intelligence technology has become an important driving force for the upgrading of the secondary industry and the realization of new industrialization, Cui Li said that decision-making artificial intelligence technology has been more applied in the field of second industries such as industrial manufacturing,

and the learning and creation of generative AI make it possible to find a new path to solve complex problems from massive data, combining generalization scenes and digital real integration.

It can greatly improve efficiency and drive industrial upgrading. Specific to the application of generative artificial intelligence in industrial fields such as manufacturing, in the basic large model,

it is necessary to combine industry data for incremental training and efficient reasoning, in order to convert industry knowledge and skills into effective output. To achieve this goal, the focus is on combining scenarios to solve the problem of data, network and computing power.

Cui Li said that China's real economy has a solid foundation, more scenes and more private data, and enterprises are actively exploring the field model of combination of size and size,

and the continuous iteration ability is constantly improving. In the secondary industry, AI will play an important role in promoting intelligent manufacturing, industrial upgrading, and new industrialization.

Cui Li said that ZTE's Binjiang manufacturing base is actively exploring the specific application scenarios of artificial intelligence technology in the production and manufacturing process,

and has achieved significant cost reduction and efficiency gains in many aspects of production and manufacturing. For example, the Nebula Industrial large model used by ZTE can shorten the product fault maintenance cycle by 30% and shorten the production scheduling time by 88%.

At the same time, based on its own practice, ZTE has also iteratively optimized a number of industry large models and embedded them into industry digitalization solutions, enabling customers to quickly deploy and achieve cost reduction and efficiency.

"Achieving replicability and commercial closed loop, with flexible universality for enterprises of different sizes, is one of the key factors for the success of generative AI in industry enablement, which requires concerted efforts across the entire industrial chain." Cui said.

Wisdom book:

Build a large model

"In 2024, the dawn of Artificial General Intelligence (AGI) dawned, with large-scale models seen as an effective way to achieve AGI." Lin Chaoting, vice president of Beijing Zhipu Huazhang Technology Co., LTD.,

said that in 2023, artificial intelligence technology has made remarkable progress, especially the development and wide application of large-scale models, which has become the focus of the global science and technology community.

However, how to promote the development of the real economy in the future AGI still faces many challenges, including the challenge of technology iteration, the challenge of commercialization,

and the challenge of safety ethics. Open cooperation and joint promotion of the development of artificial intelligence are the basis for promoting the prosperity and progress of global artificial intelligence.

Lin Chaoting introduced that Zhipu Huazhang is exploring a unique road of large-model innovation, and making progress with all parties by creating a large-model ecosystem.

First of all, Zhipu Huazhang strives to actively build a large model ecosystem. Build a layered ecology with partners from all walks of life to accelerate the application of large model innovation. Secondly,

we support open source community development. Open source and developer ecology are important because they are an important source of diversity and sustainable development of a technology.