What Happened

A new trend is emerging in the global AI market, where the importance of Central Processing Units (CPUs) is being highlighted, following the Graphics Processing Unit (GPU)-centric growth. Notably, Intel's stock price has surged by 62% since the 30th of last month, and Wall Street has signaled a dramatic revival by drastically raising its target price by 90% in one go. This comes from an analysis suggesting that CPUs are more efficient than GPUs in the AI agent phase. AMD is also showing a concurrent upward trend, driven by earnings expectations, including a contract with Meta. While NVIDIA's GPUs had virtually monopolized the AI semiconductor market for some time, CPUs are now emerging as a new core driver, signaling a shift in the overall landscape of the AI industry.

Why It Matters

The AI industry is moving beyond mere computational power enhancement and entering the era of "AI agents" applied to real life. In this process, the role of CPUs is being re-evaluated for efficiently processing vast amounts of data and performing complex tasks. While GPUs excel in parallel processing capabilities, CPUs are being assessed as more suitable for specific applications like AI agents due to their versatility and power efficiency. This shift could diversify the competitive landscape of the AI semiconductor market and reduce dependence on specific companies. Furthermore, it could be a significant inflection point that further accelerates the commercialization and diffusion of AI technology.

Impact on the Korean Market

The CPU-centric realignment of the AI market will present both new opportunities and challenges for Korean semiconductor companies. Major domestic semiconductor firms like Samsung Electronics and SK Hynix must respond sensitively to changes not only in the GPU market but also in the CPU and related High Bandwidth Memory (HBM) markets. Samsung Electronics, which operates a foundry business, could benefit from the increased demand for AI CPU production, while SK Hynix is expected to see a continuous increase in HBM demand due to the overall growth of AI semiconductors. Additionally, companies possessing essential post-processing and packaging technologies for AI semiconductor production are also anticipated to be positively impacted. Overall expectations for technological growth could inject vitality into the technology stock sector of the domestic stock market.

Key Stock Analysis

  • 005930 (Samsung Electronics): Increased demand for AI CPUs could serve as a growth driver for Samsung Electronics' foundry business. It also has the potential to indirectly boost demand for High Bandwidth Memory (HBM), contributing to improved performance in its memory business division. This will strengthen Samsung Electronics' long-term growth momentum.
  • 000660 (SK Hynix): The overall growth of the AI semiconductor market will increase demand for high-value-added memory like HBM, positively impacting SK Hynix's performance. In particular, it can further solidify its leading position in the HBM market.
  • 042700 (Hanmi Semiconductor): Demand for post-processing equipment, essential for AI semiconductor production, is expected to continuously increase, benefiting equipment manufacturers like Hanmi Semiconductor. The advancement of AI semiconductor technology promotes the growth of the equipment industry.
  • COPPER (Copper): The construction of AI semiconductors and related data center infrastructure requires enormous amounts of power and cooling systems, leading to increased demand for copper. Therefore, the growth of the AI industry could put upward pressure on copper prices.
  • BTC (Bitcoin): The advancement of AI technology can heighten overall expectations for technological growth, strengthening risk asset preference sentiment and positively impacting virtual asset markets like Bitcoin. Expectations for technological innovation can be reflected in investor sentiment.

Future Scenarios

The CPU-centric realignment of the AI market will create a new competitive landscape in the semiconductor industry. An optimistic scenario is that the revival of Intel and AMD expands the pie of the AI semiconductor market, with GPUs and CPUs developing complementarily to further broaden the application scope of AI technology. This would also provide new business opportunities for domestic semiconductor companies. Conversely, a pessimistic scenario suggests that market monopolization by companies with specific technological advantages could intensify, or unexpected bottlenecks could occur during the technology transition, leading to a slowdown in market growth. Investors should closely monitor the pace of AI agent technology development, R&D investment trends of major companies, and changes in the competitive landscape regarding performance and efficiency between GPUs and CPUs.