Since the unveiling of Chinese AI startup Moonshot AI's LLM "Kimi K3," concerns about a slowdown in AI investment and memory demand are expected to be short-lived, according to KB Securities. Kim Dong-Won, head of research, said in a report on the 22nd that algorithm optimization worries are similar to past cases with DeepSeek and TurboQuant. He argued that as AI inference expense falls, corporations will perform more AI tasks, and even if memory per inference drops to one-third, usage expanding twentyfold would increase total memory demand sevenfold. KB Securities expects Big Tech's AI investment race to continue; next year, Google and Meta's AI investments are estimated at 270 trillion won and 220 trillion won, respectively, accounting for 46% of total U.S. Big Tech AI investment. Google is expanding external data center leases and developing its "Frozen v2" server chip, while Meta plans to secure 14 GW of AI data center infrastructure by 2027. Kim projected this will boost earnings stability for SK hynix, as a higher HBM production share constrains commodity memory supply, and sales to Big Tech and AI data centers expanding to as much as 70% will ease profit volatility. KB Securities maintained a "Buy" rating on SK hynix with a target price of 4.2 million won.