News Center



Home News Center News

Practices and Collisions of Enterprise AI Intelligence

Dec 19-2024   



This event focused on the "Application of AI Intelligence in Enterprise", and the guests, alumni and students from Peking University National School of Development discussed in depth the practical implementation and challenges of AI technology in various industries, and had a wonderful discussion on the specific technical practice and future trends.

 

The Evolution of Big Models and the Collision of Enterprise Applications

Zhou Jian, founder of Lanma Technology, brought an exchange on the evolution of big models and future trends. Starting from the evolution of big models, he reviewed their role in information matching, search optimization and reducing enterprise operating costs. In particular, Mr. Zhou mentioned the construction of world models such as the intelligent simulation of complex phenomena in the real world using big models, which can help enterprises make more accurate predictions and decisions.

On the other hand, Lv Jianwei, founder of Bayesian Matrix Technology, put forward the real challenges in the application of big models from the perspective of practical application in enterprises. Mr. Lv pointed out that, especially in the Chinese market, the difficulty of data collection and the complexity of the market environment pose great challenges to the promotion of big models. He stressed that enterprises must not only rely on the technical capabilities of big models, but also face the problems of data sources, continuous optimization and feedback mechanisms.

 

The Clash between Technological Innovation and the Entrepreneurial Journey

We all know that SaaS has boomed abroad in the past decade, but SaaS in China is full of challenges. So, under the conditions of new technologies (AI big models, etc.), what can be learned from the experience of enterprise-level products?

Lv Jianwei believes that the difference between current application-level AI technology and the SaaS model is that it is highly dependent on data and real-time feedback, and that enterprise-level AI applications need to be constantly adjusted and optimized to meet changing market demands, rather than just providing generic software solutions. He believes that traditional SaaS favors generality, but the core of AI intelligence lies in adaptability and evolution. Zhou Jian believes that the deployment of AI intelligent agent should give more consideration to the deep combination with the industry, through the deep integration with the actual business scenarios, the AI technology can achieve flexible adjustment and continuous value output in the changing market environment, so as to avoid the problem of monotonous mode and easy to be replaced. And the direction of this development will lead to such applications basically becoming two systems:

                                                                     

 

But from the perspective of value creation and users, what new value can be created by the AI intelligent agent as a new form of product, and will it change the pricing of the industry?

Mr. Hou Hong, Assistant Professor at PKU NSD, has a unique perspective on the intelligent flywheel (data, model and intelligent body): This data flywheel can't actually turn up at all if it's not the same as each other. And if it doesn't turn, there will be problems in the value chain. Turning this around is exactly the logic that will allow AI intelligences to differentiate themselves from the SaaS of the past and allow business models to be established. When all three are reversed, allowing AI intelligences to generate many times the efficiency of the past, then no one can ignore the change.

With the evolution of big models and technological innovation as the core topic of this AI collision forum, the guests deeply analyzed the potential and challenges of AI technology applications in enterprises from multiple perspectives, and we look forward to more in-depth discussions on AI technology in various industries in the future.

Practices and Collisions of Enterprise AI Intelligence

Dec 19-2024   



This event focused on the "Application of AI Intelligence in Enterprise", and the guests, alumni and students from Peking University National School of Development discussed in depth the practical implementation and challenges of AI technology in various industries, and had a wonderful discussion on the specific technical practice and future trends.

 

The Evolution of Big Models and the Collision of Enterprise Applications

Zhou Jian, founder of Lanma Technology, brought an exchange on the evolution of big models and future trends. Starting from the evolution of big models, he reviewed their role in information matching, search optimization and reducing enterprise operating costs. In particular, Mr. Zhou mentioned the construction of world models such as the intelligent simulation of complex phenomena in the real world using big models, which can help enterprises make more accurate predictions and decisions.

On the other hand, Lv Jianwei, founder of Bayesian Matrix Technology, put forward the real challenges in the application of big models from the perspective of practical application in enterprises. Mr. Lv pointed out that, especially in the Chinese market, the difficulty of data collection and the complexity of the market environment pose great challenges to the promotion of big models. He stressed that enterprises must not only rely on the technical capabilities of big models, but also face the problems of data sources, continuous optimization and feedback mechanisms.

 

The Clash between Technological Innovation and the Entrepreneurial Journey

We all know that SaaS has boomed abroad in the past decade, but SaaS in China is full of challenges. So, under the conditions of new technologies (AI big models, etc.), what can be learned from the experience of enterprise-level products?

Lv Jianwei believes that the difference between current application-level AI technology and the SaaS model is that it is highly dependent on data and real-time feedback, and that enterprise-level AI applications need to be constantly adjusted and optimized to meet changing market demands, rather than just providing generic software solutions. He believes that traditional SaaS favors generality, but the core of AI intelligence lies in adaptability and evolution. Zhou Jian believes that the deployment of AI intelligent agent should give more consideration to the deep combination with the industry, through the deep integration with the actual business scenarios, the AI technology can achieve flexible adjustment and continuous value output in the changing market environment, so as to avoid the problem of monotonous mode and easy to be replaced. And the direction of this development will lead to such applications basically becoming two systems:

                                                                     

 

But from the perspective of value creation and users, what new value can be created by the AI intelligent agent as a new form of product, and will it change the pricing of the industry?

Mr. Hou Hong, Assistant Professor at PKU NSD, has a unique perspective on the intelligent flywheel (data, model and intelligent body): This data flywheel can't actually turn up at all if it's not the same as each other. And if it doesn't turn, there will be problems in the value chain. Turning this around is exactly the logic that will allow AI intelligences to differentiate themselves from the SaaS of the past and allow business models to be established. When all three are reversed, allowing AI intelligences to generate many times the efficiency of the past, then no one can ignore the change.

With the evolution of big models and technological innovation as the core topic of this AI collision forum, the guests deeply analyzed the potential and challenges of AI technology applications in enterprises from multiple perspectives, and we look forward to more in-depth discussions on AI technology in various industries in the future.