Nvidia, the dominant force in the artificial intelligence (AI) era, is expanding its influence in the personal computer (PC) market. According to U.S. media reports, Nvidia and Microsoft are set to unveil new Windows PCs that use Nvidia chips as the primary processor during events such as Computex in Taiwan. Major PC manufacturers, including Dell, are reportedly preparing to launch related products.
At first glance, this may seem like just another product announcement. However, the implications are far more significant. It signals that the stage for the AI revolution is shifting from data centers to personal desktops.
For the past two years, the AI industry has been centered around data centers. Generative AIs like ChatGPT, Gemini, and Claude have operated on massive servers. Users accessed these AIs via the internet, with responses generated in data centers connected to thousands of GPUs. Nvidia's rise to become the world's most valuable company is largely due to its near-monopoly in this market.
However, the next phase of the AI revolution is different.
AI is no longer confined to the cloud; it is beginning to enter personal PCs and smartphones. This marks the dawn of the so-called 'on-device AI' era. In this new age, AI will handle tasks such as schedule management, document creation, email organization, translation, video production, and investment analysis in real-time, making speed and security paramount. Relying solely on external servers for processing has its limitations. Ultimately, AI must operate directly on users' devices.
This is why Nvidia is venturing into the PC market.
Until now, Nvidia has been the king of AI data centers, dominating the AI training and inference market through its GPUs. Now, it aims to extend its influence into the market for end-user devices where AI is actually utilized. This is not merely about selling another chip; it is a strategy to draw the entire AI ecosystem into its platform.
In fact, Nvidia is also entering the CPU market beyond GPUs. Historically, Intel and AMD have led the CPU market while Nvidia focused on graphics cards. However, in the AI era, the lines between CPU and GPU are blurring. The efficiency of AI computations will determine competitive advantage.
Nvidia CEO Jensen Huang has a clear vision: to connect AI chips, software, data centers, cloud services, AI models, and personal devices into a single ecosystem. In simple terms, the strategy is to create a structure where “wherever AI operates, Nvidia is present.”
This is reminiscent of how Microsoft dominated the PC era with Windows, but this time the focus is on AI computing platforms rather than operating systems.
The AI revolution is now entering its second act.
If the first act was a competition among data centers, the second act will be a competition among personal AIs. In the future, people will use AI not just as a search tool but as a digital assistant. AI is likely to evolve into an 'AI agent' that understands users' schedules, emails, documents, financial assets, and health information, autonomously managing tasks.
The challenge lies in South Korea.
South Korea is a global leader in memory semiconductors. SK Hynix leads the market for high-bandwidth memory (HBM), a key component for AI semiconductors, while Samsung Electronics is also working to secure next-generation AI semiconductor competitiveness. However, in the critical areas of GPUs, AI platforms, operating systems, and large-scale AI models, U.S. companies still hold the reins.
If the current structure persists, South Korea risks remaining a parts supplier rather than a core technology provider in the AI era.
There is, however, hope. Domestic companies, including Naver, are venturing into sovereign AI development, and there is ample potential to create world-class AI applications in industries where South Korea excels, such as manufacturing, finance, healthcare, and education.
Now, the key is to choose and focus.
While the U.S. is developing AI models and China is expanding AI services based on its vast market, South Korea should concentrate on industry-specific AIs such as manufacturing AI, financial AI, healthcare AI, and defense AI. If it is a country known for making excellent automobiles, it should develop automotive AI; if it excels in shipbuilding, it should create shipbuilding AI; and if it is strong in finance, it should produce financial AI.
In the AI era, all industries will be restructured as AI industries. The time when merely producing semiconductors guarantees competitiveness is fading. National competitiveness will only be realized when semiconductors, data, AI models, platforms, and services are interconnected into a single ecosystem.
Nvidia's entry into the PC market is not just another product announcement; it marks the beginning of an era where AI transcends data centers and enters personal lives. This signifies that the competition for leadership in the AI industry has entered a new phase.
The question we now face is this: Will South Korea remain a semiconductor supplier in the AI era, or will it leap forward as a nation designing the AI ecosystem?
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.