AI PC Special Report: AI PC Industry (China) White Paper

Shared todayArtificial Intelligence SeriesIn-depth research report: "AI PC special report: AI PC Industry (China) White Paper》.

(Report produced by: IDC)

Total report: 54 pages

The historical mission of AI PC: AI is the first choice terminal for all inclusive people.

The demand for artificial intelligence is exploding, and large models have opened up a path to benefit everyone. Users not only need public large model services, but also need their own personal large models. Personal large models will rely on hybrid artificial intelligence to gradually achieve universal benefits. We see that the inclusive requirements of personal large models perfectly match the carrying advantages of PCs, indicating that PCs will once again carry the historical mission of inclusive technology and become the preferred terminal for inclusive AI.

Characteristics and inclusive requirements of personal large models

The computing load of large models continues to move from the cloud to the terminal. Public large models and local large models are mixed and used to form personal large models. Personal large models must not only inherit the powerful capabilities of public large models, but also be able to be owned by individuals and provide personalized and exclusive services to meet the various needs of users.

Four major advantages of PC carrying large personal models

Personal terminal devices include PCs, tablets, mobile phones, TVs, cars, wearable devices and other forms. However, to achieve the universal benefit of personal large models, it needs to be able to meet various requirements for the popularization of personal large models at the same time. Personal computers (PCs) have powerful computing and storage capabilities, rich interaction methods, and a wide range of application scenarios, making them an ideal platform for hosting large models.

The future definition of AI PC: Personal AI Assistant

PC has long been people's most important productivity tool and computing and interaction platform for content consumption. AI PC not only assumes the original functions of productivity tools and content consumption carriers, but also integrates hybrid AI computing units on the hardware, and can locally run "personal large models", create personalized local knowledge bases, and realize natural language interaction. , which will profoundly subvert the definition of traditional PC. AI PC is a personal AI assistant customized for each person. It not only improves production efficiency and simplifies work processes, but also better understands user preferences and protects the security of personal privacy data. It seems to be the digital expansion of the user, like the user's smart twin.

User value of AI PC

AI PC can provide users with personalized services in common scenarios, providing immediate and reliable service responses, lower costs for using large models, and reliable and secure personal data and privacy protection.

Core Features of AI PC

In order to realize the above user value, AI PC will not only be a hardware device, but a hybrid containing AI models and applications as well as hardware devices. The AI ​​PC product has a personal large model composed of a locally deployed large model and a personalized local knowledge base. The first interaction portal is a personal agent, which can realize natural language interaction. The AI ​​PC will provide hybrid information through embedded AI computing units. AI computing power can also rely on the open ecosystem to meet the needs of different scenarios. While meeting productivity improvements, personal privacy and data security are protected through local data storage and privacy and data protection protocols.

The industrial ecology of AI PC: people-oriented, terminal-led, AI-native

Driven by AI PC, the PC industry ecosystem will shift from application-oriented to people-oriented, and from application-driven to intention-driven. The traditional PC industry ecosystem is based on the operating system. Users directly operate in the system interface and manage and apply various applications. In the AI ​​PC industry ecosystem, personal agents will become the first entry point. With the support of large models and application ecosystems, they can accurately understand user instructions, give appropriate feedback, and schedule across applications to complete relatively complex tasks. Model, application, and computing power manufacturers all need to make changes around the new human-centered needs in the form of AI PC (terminal), and fully consider the efficient operation of AI in research work to adapt to the new era of AI PC.

With the continuous advancement of AI technology, more and more companies are joining the open ecosystem of AI PC, forming a prosperous multi-layered open ecosystem of users, terminal manufacturers, models, applications, and computing power. User vision will change, and the new relationship between models and applications will change the underlying logic of user needs. User needs are a key factor in promoting the development of the entire ecosystem, and terminal manufacturers integrate industry resources for users based on scenario needs to provide a hybrid delivery experience integrating software and hardware. Model manufacturers and chip manufacturers will also be more closely connected with terminal manufacturers.

Terminal Manufacturer: Advance to Ecological Organizer

Terminal manufacturers will assume the mission of organizers of the industry ecosystem, integrate industry resources for users based on scenario needs, and become the core hub of the PC industry ecosystem.

Application vendors: The rise of AI application ecosystem

The upgrade of AI PC will promote the rise of the next generation AI application ecosystem. The traditional application ecosystem is developed around operating system frameworks, providing professional business functions on top of the OS. In the new ecosystem, application development, usage methods and evaluation mechanisms will undergo disruptive changes.

Computing power manufacturer: Pratt & Whitney hybrid AI computing power

Computing power is the foundation of AI technology and the blood of the entire AI PC industry ecosystem. With the explosive development of AI terminals, computing power supply has become a key shortcoming restricting the rapid popularization of AI. In order to cope with the industry's explosive growth in demand for computing power, computing power manufacturers will carry out a series of thorough transformations to provide inclusive hybrid AI computing power as the direction of efforts and promote the comprehensive popularization of AI PCs.

AI PC market forecast: 2024 will become the first year of AI PC, accelerating PC upgrades

The evolutionary journey of China’s AI PC

The development of AI PC is a dynamic concept, with various values ​​and core features supporting and promoting each other, and gradually maturing.

China’s AI tablet and edge host market forecast

As AI PC application scenarios continue to expand, the consumer and small and medium-sized enterprise markets are expected to usher in the explosion of AI tablets and edge AI hosts. Device-edge collaboration will become the main application method of AI PC in these scenarios.

AI tablets: Tablets will also follow AI PCs in trying to run large models locally to further enhance their intelligence capabilities, especially in fields such as education and learning, leisure and entertainment, and mobile office. In terms of education and learning, AI tablets can provide a more personalized learning experience and help students better understand and absorb knowledge. In terms of leisure and entertainment, AI tablets can provide a richer entertainment experience, such as high-quality games and film and television experiences. In terms of mobile office, AI tablets can help users complete their work efficiently. It is predicted that by 2027, AI tablets will account for 75% of the tablet market.

Many families configure high-performance edge hosts as home AI hosts to provide sufficient intelligent computing power for AI terminals in the home, realize the coordination of various AI terminals and public large models in the home, and support families to create a smarter home ecosystem. . With the participation of the home host, users will be able to fine-tune the personal large model, so that the personal agent can provide personalized interactive experience and feedback optimization according to family habits and needs, and create a family-specific life butler, work assistant and tutor.

Edge AI host for small and medium-sized enterprises: Edge AI host will become a new business IT solution for small and medium-sized enterprises, and more and more small and medium-sized enterprises will deploy it. Enterprise edge AI hosts can significantly improve the edge computing power level of small and medium-sized enterprises, provide sufficient computing power support for smart offices within the enterprise, and improve work efficiency and intelligence. The edge AI host can also fine-tune the enterprise-specific model in its free time, strengthen the enterprise-specific assistant's understanding of enterprise terminology, business scenarios, and special needs, and continuously optimize the feedback effect. For small and medium-sized enterprises in different industries, edge hosts can play a differentiated role. For example, in the design industry, edge hosts can enable enterprise agents to learn from past works and design works with a consistent style.

Total report: 54 pages

Guess you like

Origin blog.csdn.net/GPT20236688/article/details/134927951