000660.KS SK hynix Inc. : Bullish and Bearish Analyst Opinions

Asian tech stocks rally on AI momentum.

Q1 earnings expectations are overly optimistic because the market is ignoring significant one-off costs, specifically employee bonuses.

The ramp of HBM4 production at a new, advanced fab positions SK Hynix to capture higher profits during the current memory super-cycle.

The speaker is constructive on SM, TSMC, Samsung, and SK Hynix, noting that Asia tech money is flowing into these AI picks and shovels companies. Rising AI demand drives semiconductor and hardware sales, benefiting these companies' financial performance and stock prices. Expectation of a grind higher supports a LONG direction. AI trade momentum could slow, or geopolitical tensions might disrupt supply chains.

The speaker states SK Hynix is "quite an attractive investment" for U.S. investors, citing its leading HBM business and its acquisition of Intel's SSD business (Solidigm), which is suitable for AI inference. The company has a unique and differentiated portfolio focused on high-growth AI segments (HBM, AI-suitable SSDs), making it a compelling play within the memory space. The explicit call for attractiveness, especially ahead of a potential U.S. ADR listing, indicates a positive view on the stock. Execution risks in integrating the Solidigm business or a loss of technological leadership in HBM.

SK Hynix is securing long-term DRAM supply agreements with prepayments from major AI hyperscalers, which reduces revenue volatility and locks in demand during an AI-driven shortage, making the stock a long.

Samsung and SK Hynix rallied ~10% and 9% in Asia on war optimism and data confirming memory chip prices are holding up, easing demand slowdown fears. The AI infrastructure buildout creates sustained demand for memory chips, creating a supply squeeze. A resolution in the Middle East removes a key overhang on tech spending. Positive price action and supportive fundamental data point to a strong near-term setup, warranting close monitoring. A re-escalation in the Middle East or a sharp downturn in global tech capex.

Author suggests rotating out of memory peers MU and SNDK to go long SK Hynix, expecting it to be the outperformer in the group.

The speaker states that Google's memory compression algorithm news has led to a "very drastic market reaction" and selling in memory chip stocks globally. She notes the trade was "one of the best performing trades in the past a year" and is "really concentrated." A technology breakthrough that reduces memory demand challenges the core growth thesis for memory chip manufacturers. The concentrated, leveraged nature of the trade exacerbates the downside momentum. The market reaction indicates the sector is now a high-risk area as a key demand driver (AI training) may be less memory-intensive than previously assumed. Analysts' calculations may prove correct that underlying demand remains extremely tight, making the sell-off an overreaction.

Google's "Turbo" algorithm can cut memory needed for LLMs by a factor of six. This news triggered a sharp, extended sell-off in Korean memory chip names like SK Hynix and Samsung. The AI-driven memory chip trade was a concentrated, high-momentum position that fueled Asia's outperformance. This technological efficiency gain introduces a perceived threat to future memory demand growth, prompting a momentum unwind. The drastic market reaction indicates elevated risk and a potential derating of these names as the "picks and shovels" AI trade for Asia comes under scrutiny. Investors should avoid due to crowded positioning and sentiment shift. Underlying physical demand for memory chips remains extremely tight, and the new technology may not immediately impact near-term demand.

The speaker states the ~10% sell-off in memory chip stocks due to Google's compression algorithm "doesn't warrant" such a move because a "big supply demand mismatch in the memory market... north of 30, 40%" persists. The new technology is an efficiency breakthrough that allows "more with less memory," which could drive greater usage and demand for AI, not reduce it. The supply shortage remains the near-term constraint. The sharp sell-off presents a potential opportunity as the negative reaction overlooks the persistent supply shortage and potential for stimulated demand. The algorithm is widely adopted by all hyperscalers rapidly, leading to a structural, permanent reduction in memory demand per unit of compute.

The tweet analyzes the strategic implications of global memory and storage capacity expansion across major industry players.

SK Hynix plans a potential US ADR listing, signaling expansion in the semiconductor market.

SK Hynix will lose its current competitive premium in memory chip manufacturing as rivals like Micron and Samsung rapidly improve their own yield curves.

SK Hynix will experience substantial margin compression because utilizing TSMC's 3nm process for their HBM4E memory is significantly more expensive than older nodes.

Speaker explicitly names Micron, Samsung, and SK Hynix as companies he loves in the high-bandwidth memory space. He states they are "cyclical companies... essentially commodities" but are currently priced at less than 10x forward earnings with high growth. Demand from AI hyperscalers is "insatiable," and these companies have sold everything they can produce through 2027, even with their announced massive CapEx. This gives them incredible pricing power and a clear multi-quarter growth runway. LONG because their oligopoly position, pricing power, and alignment with an undeniable, funded demand trend (AI infrastructure build-out) present a high-conviction opportunity, especially at current valuations. The cyclical nature of the memory business eventually leads to overcapacity and a downturn in the cycle, potentially in a few years.

The speaker maintains Sell calls on both Samsung and SK Hynix, stating they are "still seeing them to be 25%-30% above where we want their share prices to be." The view is based on expecting the memory price cycle to come off as new capacity comes online. The current high memory prices are cyclical. Significant new capacity is planned and will come to market, likely depressing prices and earnings for these dominant memory producers in the coming years. SHORT (or AVOID) based on anticipated cyclical downturn in memory prices, making current valuations unattractive despite recent pullbacks. Sustained, stronger-than-expected demand for memory (e.g., from AI servers) that outpaces new capacity additions, prolonging the cycle's peak.

Kristina stated that Nvidia CEO Jen-Hsun Huang's comment on memory optimization is a direct call out to memory suppliers like SK Hynix, Micron, and Sandisk, with Sandisk climbing higher on signals of accelerating demand. Nvidia's technology drives demand for high-bandwidth and low-power memory in data centers, benefiting these suppliers. Watch these memory companies as key beneficiaries of Nvidia's growth and increasing demand, despite current supply tightness. Supply constraints may persist, or competitive pressures could erode margins.

SK Hynix is nearing a major positive catalyst as the imminent delivery of final HBM4 samples is the last step before securing qualification and likely large orders from NVIDIA.

The stock is significantly undervalued based on 2027 consensus earnings projections from MS/Macquarie, which imply an extremely low forward P/E of 2.2x.

The author implies that Macquarie's extremely bullish revenue projections for SK Hynix are unrealistic, suggesting the stock may be overvalued based on overly optimistic growth assumptions.

The trade is a long on SK Hynix, as its projected operating income for next year is expected to exceed Apple's, suggesting the stock is still undervalued despite its significant appreciation.

A rumored delay in HBM4 volume shipments by more than a quarter could negatively impact SK Hynix's revenue and competitive positioning.

The author believes SK Hynix has further upside potential, arguing that its fundamental earnings power is underappreciated relative to mega-cap peers like Apple.

The Supreme Court ruling limits Trump's ability to impose arbitrary reciprocal tariffs, capping the new global tariff at 15% under Section 122. Asian markets (Korea +1.5%, Taiwan up) are rallying. This removes the "tail risk" of 50-100% tariffs on Asian exporters. The 15% rate is viewed as manageable and provides certainty. Furthermore, demand for AI chips remains insatiable, and these companies have pricing power to pass on the 15% cost. LONG Asian Tech Hardware. The relief rally combined with fundamental AI demand makes this the top trade. Trump finds a new legal avenue for higher tariffs after the 150-day Section 122 period expires. UUP / USD - SHORT Speaker: Mark Cranfield / Jack McIntyre Thesis: The US Dollar is trading softer following the Supreme Court decision. Jack McIntyre notes Brandywine is short USD and expects currency to be a primary return driver. The SCOTUS ruling strips the US of its "tariff cudgel," reducing US geopolitical leverage. Combined with a potential pivot to "balance of payments" tariffs (which implies the US has a deficit problem), the narrative shifts away from US exceptionalism, prompting capital flight to cheaper jurisdictions. SHORT USD. A global recession triggers a "flight to safety" back into the Dollar. GLD / SLV - LONG Speaker: Mark Cranfield Thesis: Gold is up ~2% and Silver is rallying. Traders have cut net long positions significantly in recent weeks, cleaning up positioning. The combination of a weaker US Dollar (due to the tariff ruling) and lingering geopolitical risks (Iran tensions, though secondary to tariffs) creates a perfect storm for precious metals. The "pain threshold" for tariffs is lower, meaning the Fed may not need to be as hawkish to combat tariff-induced inflation. LONG Precious Metals. A sudden spike in US real yields or a resolution to geopolitical tensions. IGV / SOFTWARE SECTOR - AVOID / SHORT Speaker: Helen Zhu Thesis: Zhu states the "AI scare trade" (selling software) is "totally justified." Companies that traded at high Price-to-Sales multiples are seeing their "terminal value" questioned. AI is disrupting the "moats" of legacy SaaS and business services. If AI can replace 80% of a service cheaper and faster, the visibility of 3-5 year software contracts evaporates. Multiples must reset lower to reflect this existential risk. AVOID / SHORT High-Multiple Software. AI adoption slows down, proving legacy software moats are more durable than expected. LATIN AMERICAN ASSETS / EMERGING MARKETS - LONG Speaker: Jack McIntyre Thesis: Brandywine Global is allocating to Latin America, citing attractive real yields and better balance sheets compared to the US. With the US Dollar weakening and US Treasuries facing supply/inflation headwinds from tariffs, capital is seeking yield in regions with positive real rates. Latin America benefits from the "weak dollar" trend. LONG Latin American Debt/Equities. A resurgence of the US Dollar or political instability in LatAm. TLT / US TREASURIES - AVOID Speaker: Jack McIntyre Thesis: McIntyre has cut net US Treasury holdings, exiting 10-year positions. He sees yields moving to the 4.5% range. Tariffs (even at 15%) are inflationary and reduce real growth, creating a stagflationary backdrop. Additionally, "bond vigilantes" may react negatively to the reduced revenue/efficiency of the US economy under a protectionist regime. AVOID US Duration. A severe US recession prompts the Fed to cut rates aggressively, sending yields lower. HONG KONG / CHINA PROPERTY SECTOR - LONG Speaker: Yvonne Man / David Ingles Thesis: The Hang Seng Property Index has seen 9 straight weeks of gains (longest run since 2003). JP Morgan upgraded the sector to overweight. There is "FOMO sentiment" from mainland Chinese buyers diversifying wealth into Hong Kong assets. The removal of extreme tariff risks (fentanyl-related tariffs gone, replaced by flat 15%) improves the macro outlook for Chinese wealth preservation. LONG Hong Kong Developers/REITs. Regulatory crackdowns from Beijing or a resurgence of US-China tensions post-summit. TEXTILES & LEATHER / CONSUMER DISCRETIONARY (China Exporters) - LONG Speaker: Yvonne Man / Matt Priest Thesis: "Shadow International" (Shenzhou International) is up 4%. Matt Priest notes that a flat 15% tariff is better than the chaos of reciprocal levies (which were 20-100%). The market feared a trade war escalation with tariffs hitting 60-100%. A capped 15% rate is a "relief" for exporters who have already diversified supply chains or have pricing power. The uncertainty premium is being priced out. LONG Major Asian Exporters. Trump uses other tools (Section 301/232) to target specific sectors after the 150-day window.

A CLSA research note projects SK Hynix will maintain its dominant leadership position in the next-generation HBM4 memory market with a 55% share.

Long SK Hynix (specifically the European listing) as a statistical arbitrage play when the US-listed South Korea ETF (EWY) rises significantly, expecting the heavily-weighted component to catch up.

AI lab compute spend is growing rapidly, and the differentiated memory market is an oligopoly trading at cyclical multiples, insulating it from broader macro risks.

1. THE FACT: The speaker asserts that over the next 5-10 years, demand for power, chips, and memory, driven by AI, will massively outstrip supply. Companies like GE Vernova (GEV), TSMC (TSM), SK Hynix (000660.KS), and Samsung (005930.KS) are cited as having fully packed order books. 2. THE BRIDGE: This structural supply/demand imbalance creates a powerful, long-term tailwind for these key AI infrastructure and supply chain companies. 3. THE VERDICT: Establish long positions in core AI infrastructure providers across power (GEV) and semiconductors (TSM, SK Hynix, Samsung) to capitalize on the multi-year AI boom.