Español
Português
Việt Nam
日本語
русский 
عربي 
Deutsch
한국어
Türkçe
Français
Italiano
On October 24, Lenovo Group unveiled its most comprehensive artificial intelligence product technology to date, and its vision of "AI for All" at the Tech World. This year's event, Lenovo's ninth, focuses on how Lenovo is unleashing the power of AI to drive intelligent transformation in every aspect of people's lives and every industry - from the pocket to the cloud, providing dedicated AI devices, infrastructure, solutions and services to industries, enterprises and individuals around the world.
Yang Yuanqing, chairman and CEO of Lenovo Group, said that artificial intelligence technology has become so real for the first time that it is closely related to every person and every enterprise. While rejoicing in technological progress, we have also begun a deeper exploration.
Yang Yuanqing, CEO of Lenovo Group: For the first time, artificial intelligence technology has become so real
What are the "key but invisible" elements that make generative AI and large language models work? There is no doubt that data, as a product of the first phase of digital transformation, is essential. Algorithm/model-based training and reasoning are key, but at the same time, computing power is indispensable, not only on the cloud, but also with more powerful computing power on the edge side and the end side.
To create panoramic AI, "pocket to cloud" computing power, multiple forms of application are needed, and solutions from different industries are also critical.
There is a dilemma here, if the user wants the public big model to answer your questions or talk to you in the right and appropriate content that you want, then you have to honestly tell it what you really think, the real records, the data, but then the user's personal data and even the privacy and the business secrets become part of the public information.
People want the ability to have big models that can answer questions, but they also want their data to live only on their own devices or within their own businesses. How can you do "both want - also want"? The answer is obviously not limited to public large models, we can achieve through individual large models and enterprise large models.
Lenovo's large model compression technology enables our own smart terminals and devices to run personal-scale large models. On these AI-enabled terminals and edge devices, local knowledge bases will be built to better understand users. The individual large scale model will use personal data stored on a device or home server for inference. Unless authorized by the user, the user's personal data is never shared or sent to the public cloud, thus ensuring personal privacy and data security.
The enterprise large model will coexist with the public large model and public cloud to form a hybrid form and hybrid deployment of artificial intelligence. This hybrid trend is already occurring in many scenarios, such as hybrid and digital office Spaces.