Samsung Selloff a Buying Opportunity Amid Looming Chip Shortage, Says Analyst

A recent 30% decline in Samsung Electronics' share price from its peak presents a long-term buying opportunity, as the drop reflects weakening investor sentiment rather than a deterioration in the company's fundamentals, according to an analyst note from KB Securities.

Analyst View: Selloff Detached From Fundamentals

In a note to investors, KB Securities analyst Jeff Kim addressed the stock's recent performance, which has been fueled by concerns over a potential slowdown in artificial intelligence investments. The firm stated that the price correction appears driven by "overblown concerns rather than damage to fundamentals."

According to Kim, the core market drivers remain intact. The note emphasized that "both long-term growth outlooks for the AI infrastructure sector and the key industry fundamental (chip shortage) remain unchanged from a month ago."

The Coming Chip Shortage

KB Securities projects that the industry is heading for a severe supply crunch, forecasting that the chip shortage in 2027 could be "the worst in the semiconductor industry’s 70-year history" and will likely persist through at least 2028.

Ad

This expected shortage is attributed to aggressive capital expenditure focused on high-demand High-Bandwidth Memory (HBM), which structurally limits the industry's ability to increase output capacity for commodity DRAM.

• The share of HBM in global DRAM wafer output is expected to rise from 15% in 2026 to 34% in 2027.
• This means new manufacturing capacity will be largely allocated to HBM, tightening the supply of other memory types.

AI Data Center Boom a Key Catalyst

A primary driver for this sustained demand is the accelerating build-out of AI data centers in the United States, the note said. The U.S. Federal Energy Regulatory Commission has reportedly fast-tracked grid connection procedures, shortening lead times from five years down to one to two years.

This regulatory change is expected to double the pace of AI data center deployment. As an example of the scale of this expansion, the analyst cited Meta’s plans to add 14GW of AI computing infrastructure by 2027.