KB Securities said in a note distributed Wednesday that the recent selloff in a leading memory company presents what it views as a long-term buying opportunity. The brokerage's analyst, Jeff Kim, argued the stock's pullback reflects fading investor confidence rather than any meaningful deterioration in the sector's underlying fundamentals.
Kim pointed to Samsung Electronics, noting the stock is down roughly 30% from its prior peak amid market concerns about a slowdown in AI-related investment. KB Securities, however, maintained that both the long-term growth trajectory for the AI infrastructure market and the core industry dynamic - a looming chip shortage - remain unchanged compared with a month earlier.
The firm characterized the recent price correction as being driven by "overblown concerns rather than damage to fundamentals." KB Securities additionally forecast that the chip shortage expected in 2027 will be the worst in the semiconductor industry's 70-year history, and that supply tightness is likely to persist through at least 2028.
KB attributed much of the future shortfall to aggressive capital expenditure focused on HBM. The firm said this HBM-centric capex will structurally constrain the ability of commodity DRAM production to expand, because a growing share of new wafer output will be allocated to HBM rather than to commodity memory.
According to Kim's estimates cited by KB Securities, HBM's share of global DRAM wafer output should increase from about 15% in 2026 to roughly 34% in 2027. That shift implies that a substantial portion of new capacity additions will serve HBM needs, leaving less incremental capacity available for commodity DRAM.
On the demand side, Kim highlighted an acceleration in AI data center construction across the United States. He noted that the Federal Energy Regulatory Commission has shortened grid connection lead times by fast-tracking procedures, reducing typical waits from five years to between one and two years.
KB Securities said the shortened connection timelines should double the pace of AI data center deployment. The analyst pointed to Meta's plan to add 14 GW of AI computing infrastructure by 2027 as an example of the scale of demand underpinning the memory market.
This assessment from KB Securities frames the recent share price weakness as a potential entry point for long-term investors, grounded in expectations of persistent supply constraints and accelerating AI-driven demand that together support near- to medium-term pricing power for memory chips.