In 2024, the PC market seems to be radiating new vitality and vitality, driven by the recovery of demand and AI PCs!
Recently, a number of market research institutions released the 2023 global PC market shipment report. According to data from Canalys, for the whole year of 2023, the total number of PC market shipments will be 24.7 billion units, down 13% year-on-year; Data from Counterpoint Research also shows that in 2023, shipments in the global PC market will decline by 14% year-on-year. This shows that PC manufacturers are still in the middle of a cold winter in 2023.
But the good news is that the global PC market ushered in an inflection point in the fourth quarter of 2023, showing signs of recovery. According to data from Canalys, in Q4 2023, shipments in the global PC market increased by 3% year-on-year, ending seven consecutive quarters of year-on-year decline. Counterpoint said that in Q4 2023, the global PC market shipments will decrease by 02%, which is expected to usher in the inflection point of the shipment contraction cycle.
Market analysis points out that the global PC market has reached the bottom and is expected to start to grow in 2024, and AI PC is expected to be an important driving force for the growth of the PC market under the huge wave set off by generative AI.
PC manufacturers clench their AI thighs
In the past few years, the traditional PC industry has been shrouded in the haze of sluggish sales, but entering 2024, the industry generally believes that after a long-term adjustment, the PC market is expected to return to the growth track this year.
The improvement of market expectations for the PC industry is mainly based on several considerations。The first is that business and consumer confidence is likely to strengthen as global economic activity resumes, leading to a recovery in PC demand. Secondly, the inventory levels of manufacturers such as Lenovo Group, HP, and Dell have returned to normal and are close to healthy levels.
More importantly, with the boom of generative AI, the industry generally believes that AI PCs are expected to rekindle new hopes in the PC market and start the first year of AI PCs in 2024.
From the demand side, AI PCs, as devices that carry large AI models, can effectively improve productivity, and it will become necessary to upgrade from traditional PCs to AI PCs. In addition, after the popularization of AI, the importance of AI localization has become more and more prominent due to privacy and cost considerations.
Canalys notes that initial adoption of AI PCs will be limited to specific business functions and tech-savvy power users who need powerful hardware to run intensive AI workloads. But over the next five years, demand for AI PCs will surge, driven by the emergence of a market for fine-tuned large language models (LLMs) and AI tools, as well as modified operating systems defined by powerful AI integrations.
Previously, Intel CEO Pat Gelsinger told ** that AI PC is a key turning point in the PC market in the next few years. Nvidia CEO Jensen Huang also believes that the computer industry is ushering in a regeneration opportunity, and in the next decade, new AI computers will replace trillions of dollars worth of traditional computers.
According to Canalys**, AI PCs will account for nearly 20% of PC shipments in 2024. By 2027, this proportion will quickly grow to more than 60% thanks to variable energy and new local experiences.
In order to grasp the development trend of AI PCs, giants have left the market. Since the beginning of 2024, AI PC has ushered in a number of blockbuster news.
Earlier this month, Microsoft announced a major change to the Windows PC key layout, introducing the Copilot key, where users will be able to call up Microsoft's AI (artificial intelligence) assistant with one click, and the first devices with the new keys will be available this month.
In addition, at the recent CES exhibition, AI PC was one of the biggest focuses. The day before the CES opening, Nvidia unveiled its latest GeForce RTX 40 series graphics cards; At the conference, Intel announced that the Arrow Lake processor and Lunar Lake processor, which focus on AI performance, will be available in 2024.
In the field of terminal products, Lenovo Group, the world's leading PC company, is also accelerating the advancement of AI PCs. It is understood that at present, Lenovo Group has taken the lead in entering the "AI ready" stage, and has sold ThinkPad X1 Carbon, Xiaoxin Pro16AI Core Edition and other related new products. This year, it is also expected to launch its new "AI On" product equipped with a large personal model.
In addition, brand manufacturers such as Dell, HP, and Asus have also announced their own AI PC plans. Many scenes seem to herald the arrival of a new era of AI PC.
"Chunjiang Plumbing Duck Prophet", in the secondary market, AI PC concept stocks have been warmed up in advance. For example, the global PC leader Lenovo Group (00992HK) has continued to rise in the past year, soaring by more than 8% throughout the year, becoming the best performer among the constituent stocks of the Hang Seng Index in the same period.
Real AI PC or not yet available?
The AI PC story is good, but the rise of a new industry is not an overnight process, but a gradual process.
IDC's first "AI PC Industry (China)**" released a few days ago clarifies that the development of AI PC is a continuous evolution process, which should be divided into AI ready stage and AI ON stage.
According to the technical roadmap disclosed by Lenovo Group, in the AI Ready stage, the main focus is on the upgrade of the chip computing architecture, so that the AI PC has high local AI inference computing power; In the AI ON stage, the AI PC will realize the integration of software and hardware ecosystems, embed the local AI large model knowledge base and rich applications, and the ecosystem has been basically improved.
In the view of Ablikim Ablimiti (Amu), vice president of Lenovo Group and vice president of strategy and business development in ChinaA mature AI PC should have five characteristics. First, there should be a personal agent based on a large model, with which users can interact naturally; The second is to configure personal local models and knowledge bases, which can be combined into personal models; the third is to configure hybrid AI computing power; Fourth, it can connect to AI native applications through plug-ins; Fifth, there are privacy and data security protection programs.
It can be seen that whether a large model can be installed on the PC and the fine-tuning training of the individual large model can be realized in the edge private domain environment has become the key factor to judge the maturity stage of the AI PC.
However, to put a large model into a PC, the AI PC will definitely face many challenges. First of all, the implementation of AI PCs requires powerful computing power, ranging from tens of billions of parameters to hundreds of billions of parameters. This has always been the focus of research for chip manufacturers such as Intel, AMD, and NVIDIA.
Secondly, at present, users are using public large models in the cloud, and large models should be able to achieve "personalization" for individuals, and having local computing power is almost an inevitable condition for AI PCs. Furthermore, data security and privacy are the bottom line premises.
Conclusion:
To sum up, the explosion of AI PC is the general trend, which will further open up the imagination space for the industry. In order to compete for the first opportunity in the layout of AI PCs, giants such as Intel, AMD, Qualcomm, Lenovo, Dell, and HP have begun to fight with real guns. For them, it's a battle that could change their status.
However, although the AI PC has stepped onto the stage of history, its development is still uncertain, and its future is still full of many uncertainties and questions. Standing at the starting point of the AI PC era, who can carry the banner of AI PC first in the future? Everything is still unknown.
Author: Bottle.