Journalist
Candice Kim김혜준
candicekim1121@ajupress.com
ReporterSamsung Electronics, SK hynix, LG Electronics, Olive Young, Musinsa & Semiconductor, K-Beauty
Candice Kim is a dedicated business and technology reporter specializing in the intersection of high-tech innovation and fast-paced consumer trends. Her core beats span the semiconductor, IT, cosmetics, and retail sectors, where she provides in-depth coverage of industry titans including Samsung Electronics, SK hynix, and LG Electronics, alongside leading consumer brands like Olive Young and Musinsa. Whether analyzing complex silicon supply chains or tracking the latest shifts in beauty and fashion retail, Candice is committed to delivering sharp, accurate reporting that keeps readers informed on the forces shaping today's most dynamic markets.
"Connecting the dots from silicon chips to consumer shifts."
Latest by Candice Kim
-
LG Electronics brings next-gen 5G to European cars in SDV push SEOUL, August 19 (AJP) - LG Electronics has begun supplying its latest 5G telematics technology for a mass-produced vehicle in Europe, marking the region's first deployment of the R16 standard as the Korean electronics maker strengthens its position in the shift toward software- and AI-defined vehicles. The company said Wednesday its new smart telematics solution has been installed in a European automaker's production model, though it did not identify the customer or vehicle. The system is based on 5G Release 16, or R16, the second major 5G standard designed to improve reliability and reduce latency over the earlier R15 specification. LG said the system achieves a data error rate of 0.001 percent and transmission latency of 1 millisecond between the telematics module and base station, representing improvements of 100 times and five times, respectively, over 4G technology. Those capabilities are becoming increasingly important as vehicles evolve into software-defined vehicles, or SDVs, and AI-defined vehicles, or AIDVs, which require large volumes of data to be exchanged quickly and reliably. The technology supports over-the-air software updates, vehicle-to-everything, or V2X, communication and large-scale data transmission for autonomous driving systems. Lower latency and higher reliability are particularly important for autonomous driving, where vehicles must rapidly exchange information with surrounding infrastructure and other devices. LG has also integrated the communications module and as many as 12 antennas into a single unit, replacing conventional systems in which antennas are distributed across different parts of a vehicle. The company said the design reduces signal loss, wiring and installation space while potentially lowering costs for automakers. The rollout builds on LG's existing position in automotive telematics, where it holds the largest global market share based on estimates from TechInsights. The company has been expanding its automotive portfolio beyond infotainment into connectivity, in-cabin sensing and software platforms as vehicles become increasingly software-driven. LG also joined the Global 6G Coalition in March to participate in standards development and system verification as the automotive industry begins preparing for the next generation of wireless connectivity. "We will continue to introduce innovative solutions in automotive communications, which has become a key competitive area in mobility, and further strengthen our telematics market leadership in the transition from SDVs to AIDVs," said Eun Seok-hyun, head of LG Electronics' Vehicle Solution business. AJP Takeaways LG has deployed 5G R16 telematics in a mass-produced European vehicle for the first time in the region, bringing lower latency and greater data reliability to connected cars. The system combines a communications module with up to 12 antennas and supports OTA updates, V2X communication and data-intensive autonomous-driving functions. The rollout strengthens LG's push beyond automotive infotainment into the connectivity infrastructure needed for software- and AI-defined vehicles. 2026-08-19 10:13:23 -
Samsung bets $173 million on Korea HVAC line as AI data-center cooling demand heats up SEOUL, August 19 (AJP) - Samsung Electronics is stepping up its bet on AI data-center cooling with a 240 billion won ($173 million) production line in South Korea, targeting growing demand to manage the heat generated by increasingly powerful AI servers. The company said Wednesday it will build the new heating, ventilation and air conditioning (HVAC) production line at its Gwangju complex, expanding its push into AI data-center cooling following its acquisition of FläktGroup last year. Samsung, FläktGroup Korea and the local government signed an investment agreement Wednesday for the facility at Samsung's Gwangju manufacturing complex. The new line will span 21,800 square meters, roughly the size of three soccer fields, and is scheduled to begin operations in early 2028. The facility will manufacture coolant distribution units, or CDUs, a key component of liquid-cooling systems used in AI data centers, as well as fan wall units, computer room air handlers and air handling units. Demand for liquid cooling has been rising as increasingly powerful AI servers generate more heat, putting pressure on conventional air-cooling systems. CDUs regulate the temperature and pressure of coolant circulating between a data center's chiller and cold plates attached to servers. The Gwangju line will become FläktGroup's 15th production facility globally. The HVAC specialist currently operates 14 production sites and eight research centers across the United States, Europe and Asia, supplying equipment for data centers, industrial plants and commercial buildings. The investment marks Samsung's first major expansion of production facilities at its Gwangju complex since the site was established in 1989. The plant currently produces appliances including refrigerators, washer-dryers and air conditioners. Samsung plans to combine FläktGroup's central HVAC technology with its SmartThings Pro connectivity platform and building management systems as it expands sales to AI data centers and advanced manufacturing facilities. "Samsung Electronics will rapidly expand HVAC as a future growth engine and seek to become a leading player in the global market," said Kim Cheol-gi, head of Samsung's Digital Appliances business. "The new Gwangju production line will provide an important foundation for strengthening our competitiveness in HVAC." AJP Takeaways Samsung will invest 240 billion won in a new Gwangju HVAC production line scheduled to begin operations in early 2028. The facility will produce CDUs used in liquid cooling for AI data centers, turning Samsung's FläktGroup acquisition into additional manufacturing capacity in Korea. The investment extends Samsung's AI infrastructure push beyond semiconductors, targeting the rapidly growing cooling requirements of increasingly power-dense AI data centers. 2026-08-19 09:35:22 -
China cracks memory club, but harder test looms SEOUL, August 14 (AJP) - China's heavily government-backed chipmakers are breaking into the once-exclusive global memory club, with Yangtze Memory Technologies Co. (YMTC) vaulting into the world's top three NAND flash suppliers for the first time. But just as China closes the gap in manufacturing scale, artificial intelligence is moving the industry's profit pool toward a harder target. YMTC captured 14 percent of global NAND bit shipments in the second quarter, overtaking Japan's Kioxia to rank behind Samsung Electronics at 25 percent and SK hynix at 22 percent, according to Counterpoint Research. The ranking marks a striking ascent for a company founded only a decade ago and offers some of the clearest evidence yet that China's push for semiconductor self-sufficiency is reshaping a memory market long dominated by a small group of South Korean, Japanese and U.S. suppliers. Much as ChangXin Memory Technologies (CXMT) has broken into the DRAM market, YMTC has achieved scale that is yet to translate into meaningful numbers on the top and bottom lines. It ranked third in NAND shipments during the quarter but only fifth in revenue, behind Kioxia and Micron Technology, as its product mix remained concentrated in lower-priced consumer applications rather than high-value data-center storage, Counterpoint said. That distinction matters more than ever. Enterprise solid-state drives, or eSSDs, accounted for 48 percent of global NAND bits shipped in the second quarter, up sharply from 26 percent a year earlier. Counterpoint expects their share to exceed half of all NAND shipments by the end of this year. The shift reflects the widening deployment of AI from model training toward inference, where enormous volumes of data must be stored and retrieved rapidly and with low power consumption. That is turning storage — long overshadowed by graphics processors and high-bandwidth memory in the AI investment boom — into an increasingly important part of data-center infrastructure. For Samsung and SK hynix, it creates an unusual competitive dynamic. YMTC's rapid expansion threatens to intensify competition in mainstream NAND, where additional Chinese supply could eventually pressure market share and pricing. At the same time, the industry's most valuable demand is migrating toward enterprise products, giving the Korean companies another layer of defense. Lee Jong-hwan, a professor of system semiconductor engineering at Sangmyung University, said YMTC's rise should not be dismissed as a temporary phenomenon driven simply by price competition. "Chinese memory companies are different from what they used to be," Lee said. "Their technological capabilities have improved considerably, and they are investing heavily in research and development at a rapid pace to compete with Samsung and SK hynix." Samsung has been expanding high-capacity enterprise SSD products for AI servers, while SK hynix has built a substantial data-center storage business through Solidigm, the U.S.-based company created after its acquisition of Intel's NAND and SSD operations. SK hynix's share in the second quarter was helped by a roughly 40 percent quarter-on-quarter increase in Solidigm's bit shipments, according to Counterpoint. YMTC increased shipments 22 percent from a year earlier and 5 percent from the previous quarter as shortages emerged in parts of the consumer market while established suppliers increasingly favored higher-margin products. China, in other words, is catching up fastest where the market has been. The harder question is whether it can catch up where the market is going. For YMTC, sustaining its new ranking will increasingly require competing in enterprise SSDs, where it faces higher barriers than in supplying NAND for smartphones and personal computers. Data-center customers demand endurance, reliability, power efficiency and sustained performance, while products typically undergo extensive qualification before they can be deployed across servers at scale. Samsung, SK hynix and Micron also have long-established relationships with global cloud and server customers that cannot be replicated simply by adding fabrication capacity. Like other Chinese names, YMTC has been on the U.S. Commerce Department's Entity List since December 2022, subjecting exports, reexports and transfers of covered U.S. technology to stringent licensing requirements. Those restrictions have constrained its access to certain advanced chipmaking technologies and added another commercial and compliance hurdle as it seeks to expand internationally. Washington's broader semiconductor controls have simultaneously reinforced Beijing's determination to build a more self-sufficient chip industry, turning memory into one of the central battlegrounds in China's technology push. Still, the restrictions have not stopped YMTC's technological or capacity advance. The company is mass-producing 267-layer 3D NAND and developing technology with more than 300 layers using its Xtacking architecture, while continuing to expand supply to Chinese device makers, according to Counterpoint. YMTC has also begun moving toward enterprise SSDs, the research firm said, with plans to increase their share in its product mix during the second half. The stakes are particularly high because NAND has become the second front of China's challenge to the global memory hierarchy. CXMT has already emerged as the No. 4 DRAM player and made a blockbuster Shanghai debut on July 27 after raising 57.92 billion yuan ($8.6 billion) in the biggest semiconductor IPO ever on mainland China. Its shares surged 466 percent on their first trading day. YMTC is set to follow the same path. The NAND maker formally began the process for a listing on Shanghai's technology-focused STAR Market in May, potentially opening another source of capital for capacity expansion and next-generation development. State-backed entities, including China's national semiconductor funds and local government investors, are among its major shareholders. Together, CXMT and YMTC represent something the global memory industry has not previously confronted at this scale: Chinese challengers gaining ground simultaneously in both of its major markets. The threat is not yet symmetrical. In DRAM, Samsung and SK hynix retain a formidable lead in high-bandwidth memory, or HBM, the premium memory stacked alongside AI accelerators. In NAND, the Korean companies similarly hold stronger positions in high-value enterprise storage even as YMTC closes the shipment gap. But memory-chip history shows why scale itself matters. The industry is notoriously cyclical. Relatively modest changes in supply can produce sharp swings in prices once inventories start accumulating. A well-funded Chinese supplier capable of steadily adding capacity could therefore influence the broader market long before it catches the incumbents at the technological frontier. Lee said that is precisely where YMTC's growing scale could become a more immediate concern for the Korean memory makers, even before it closes the technology gap. "The larger their share becomes, the greater the possibility that prices will come down," Lee said. "Samsung and SK hynix will have to keep investing and pursuing research and development to maintain the gap." For Samsung and SK hynix, the challenge is consequently no longer simply to keep YMTC behind them in shipment rankings. As with smartphones, Korean chipmakers are likely to separate themselves by concentrating on the premium market — HBM in DRAM and enterprise SSDs in NAND. YMTC has crossed the first barrier, and its ability to cross the next will determine if China's NAND rise remains largely a volume story — or becomes a direct challenge to the memory industry's most profitable markets. 2026-08-14 16:55:37 -
Samsung brings AI brains as chip rivalry with SK hynix heats up SEOUL, August 14 (AJP) - Samsung Electronics has recruited two artificial intelligence specialists to accelerate the use of AI in semiconductor research and manufacturing, as its intensifying rivalry with SK hynix expands beyond advanced memory chips into how those chips are designed and produced. Samsung has hired Seoul National University professor Bohyung Han, a specialist in deep learning and computer vision, as a Fellow to lead the development of AI models tailored to semiconductor research and development, according to information provided by the company. Tairin Hahn, formerly a senior engineering manager at Meta, has joined as vice president and will oversee the development of "AI-ready data," or datasets structured for direct use by AI systems. The appointments come as Samsung seeks to apply AI more extensively across chip design, process engineering and manufacturing, areas where increasingly complex semiconductor development is generating vast amounts of data and placing greater demands on engineering productivity. Samsung has already laid out plans to turn its manufacturing operations into AI-driven factories by 2030, using AI agents and digital twins across production, quality control and logistics. In its semiconductor operations, the company has also recreated its Pyeongtaek fab as a digital twin using Nvidia's Omniverse platform. Rival SK hynix is pursuing a similar transformation. The HBM market leader is targeting autonomous semiconductor manufacturing by 2030 and has expanded its partnership with Nvidia to use AI and digital twins in chip design and production. The parallel push suggests that competition between South Korea's two memory giants is increasingly extending beyond HBM performance and production capacity into the use of AI to shorten development cycles, improve yields and extract more output from increasingly expensive fabs. AJP Takeaways: Samsung is adding specialized AI talent to semiconductor R&D and manufacturing, hiring Bohyung Han to develop chip-focused AI models and Tairin Hahn to build data infrastructure that those models can readily use. The hires fit into Samsung's broader push to turn its manufacturing operations into AI-driven factories by 2030, applying AI across chip design, process engineering and production. The move opens another front in Samsung's rivalry with SK hynix, as the two Korean memory giants increasingly compete not only in HBM technology and capacity but also in using AI and digital twins to shorten development cycles and improve manufacturing efficiency. 2026-08-14 14:09:42 -
LG to roll out Nvidia-powered humanoid Q1 2027 SEOUL, August 14 (AJP) -South Korea's LG Group plans to unveil a bipedal humanoid robot powered by Nvidia's artificial intelligence technology in the first quarter of next year, pushing partnership beyond technology and component supply into joint development and commercialization across robotics, AI factories and autonomous driving. The electronics giant said Friday it signed a memorandum of understanding with Nvidia at the U.S. chipmaker's headquarters in Santa Clara, California, on Thursday, following talks between LG Chairman Koo Kwang-mo and Nvidia CEO Jensen Huang in Seoul in June. The agreement puts concrete timelines behind that partnership, with the companies targeting a humanoid debut in early 2027, an Nvidia Vera Rubin-based AI factory reference site later that year and an 80-megawatt facility in Cheonan in 2028. The first major project will be a next-generation bipedal humanoid built on Nvidia's Isaac GR00T open foundation model. The robot will use Nvidia's Jetson Thor computing platform and Halos for Robotics safety system. LG will contribute hardware from across its affiliates, including actuators from LG Electronics, sensors from LG Innotek and batteries from LG Energy Solution. The group aims to unveil the robot in the first quarter of 2027. Before then, LG Electronics plans to deploy its wheeled CLOiD robot on a washing machine production line at its Tennessee factory by the end of this year for proof-of-concept testing, taking its robotics push from household demonstrations into an actual manufacturing environment. LG and Nvidia will also work on a physical AI data factory using LG CNS' PhysicalWorks robotics platform, creating a system for collecting, generating, training and validating robot data. Data gathered from factory deployments will feed back into a robot foundation model that LG is developing in-house. The partnership extends into the infrastructure needed to train and operate those AI systems. LG plans to build a reference AI factory in the first half of 2027 using Nvidia's next-generation Vera Rubin platform and DSX architecture to test its cooling, power and IT technologies. The project will expand into an 80-megawatt AI factory in Cheonan, South Chungcheong Province, in the first half of 2028. LG said the facility will use prefabricated modular construction, with servers, cooling equipment and power systems assembled in modules before installation on site, cutting construction time by more than 20 percent. LG plans to combine cooling technology from LG Electronics, batteries from LG Energy Solution, the data-center design and operating capabilities of LG CNS and LG Uplus, and power solutions from LS. It ultimately aims to package those technologies into integrated AI infrastructure that can be offered to global technology companies. The Cheonan facility will also serve as an operating base for physical AI and the development of robot foundation models, linking LG's infrastructure investment directly with its robotics ambitions. In mobility, the companies will develop a high-performance computing platform for AI-defined vehicles based on Nvidia's Drive Hyperion architecture and LG's in-vehicle infotainment and automotive software technologies. The project would push LG's automotive electronics business beyond its traditional strength in infotainment and further into autonomous driving and in-vehicle AI, opening another potential route to global automakers. The companies will establish a joint task force covering research and development, field testing and commercialization. "As our collaboration gains speed, the areas where we can work together in AI factories, physical AI and mobility have become clear," Koo said. "We will accelerate the adoption of AI by building industry-leading reference cases." Huang said physical AI could give machines the ability to understand and reason about the real world and operate safely alongside people. "LG and Nvidia are combining LG's leadership in product engineering and manufacturing with Nvidia's technology to usher in a new era of robotics, AI factories and autonomous vehicles," Huang said. Shares of LG Electronics rose 2.2 percent to 210,000 won ($419) as of 11:10 a.m. Friday. AJP Takeaways: LG targets the first quarter of 2027 to unveil a bipedal humanoid using Nvidia's Isaac GR00T, Jetson Thor and robotics safety technologies. The alliance stretches from robots into infrastructure, with a Vera Rubin reference site planned for 2027 and an 80 MW Cheonan AI factory in 2028. LG is using the Nvidia partnership to connect businesses spanning electronics, batteries, IT services and automotive components into a broader physical-AI platform. 2026-08-14 11:04:02 -
Korea takes AI race from chips to chillers SEOUL, August 13 (AJP) - An AI agent on a laptop or smartphone may appear almost weightless. Behind every prompt, however, sit warehouses of processors and memory running around the clock — consuming enormous amounts of electricity and generating heat that must be removed before the machines can keep thinking. For South Korea, already a powerhouse in the memory chips feeding those machines, that physical burden is becoming the next AI opportunity. Samsung Electronics has already put money behind that opportunity, betting that cooling will become a major business alongside the chips powering AI. The company put that strategy on display Thursday with the completion of a new production line for FläktGroup, its heating, ventilation and air-conditioning business, in Pune, India, as data-center investment accelerates across India and the wider Asia-Pacific region. The 13,826-square-meter facility can produce as many as 6,500 HVAC units a year when fully operational, including air handling units, fan wall units and computer room air handlers used to manage temperature and airflow in data centers and large commercial buildings. The plant reached mass-production readiness about six months after equipment installation began early this year. FläktGroup said in May it was investing 5.7 million euros ($6.7 million) in the Pune facility, with more than 12,000 square meters devoted to production. Pune was chosen partly for its location near industrial corridors and ports serving one of India's fastest-growing data-center markets. The expansion comes nine months after Samsung completed its acquisition of FläktGroup, Europe's largest HVAC company, giving the Korean technology giant a foothold in a business becoming increasingly important to AI infrastructure. Samsung agreed in May 2025 to acquire the company for 1.5 billion euros and completed the transaction in November. Samsung said when it announced the deal that the global data-center cooling market was projected to grow about 18 percent annually through 2030, significantly faster than the broader applied HVAC market. The bet reflects a broader change in the economics of AI. As more powerful graphics processors and other accelerators are packed into server racks, the infrastructure surrounding those chips — memory, storage, electricity and cooling — is becoming a constraint on how quickly additional AI computing capacity can be deployed. Moon Song-chun, professor emeritus at the Korea Advanced Institute of Science and Technology, said the growing importance of cooling stems from the enormous increase in computing and data-processing power required by AI workloads. "Without proper cooling, a data center simply cannot process the data," Moon told AJP. "Cooling technology is one of the most critical elements." AI data centers are not fundamentally separate from conventional data centers, Moon said. What has changed is the intensity of computing. Vast clusters of GPUs process far larger workloads while users still expect rapid responses, putting heavier demands on electricity supply and thermal management. The technology is already forcing a change in how servers are cooled. TrendForce estimates that liquid-cooling penetration in AI server racks will reach 47 percent in 2026 as the thermal design power of individual AI processors rises above 1,000 watts. The research firm says next-generation AI architectures are moving liquid cooling from an optional configuration toward an essential specification. For Korea, that creates an opportunity to stretch an existing advantage in AI memory across more of the infrastructure stack. SK links memory with data centers SK Group is pursuing that integration from another direction, seeking to connect memory production with the data centers that consume it. SK Telecom last month outlined plans for as much as 15 gigawatts of AI data-center capacity in South Korea, with the goal of turning the country into an Asian AI infrastructure hub. The company has since established SK Hyper, a dedicated data-center development subsidiary, and committed 750 billion won through 2030 for groundwork including site acquisition and substations. The scale illustrates how capital-intensive the next stage of AI could become. SK Telecom estimates that building a typical 1-GW AI data center could require about 70 trillion won because of the computing hardware, memory and supporting infrastructure involved. The group plans to combine its own capital with strategic investors, long-term customer contracts and project financing. SK Telecom is also working with Nvidia on a gigawatt-scale AI cloud in Korea using Nvidia's DSX architecture, with the first AI factory planned to begin operations in 2027. That data-center push gives SK an opportunity to link computing infrastructure with another business in which the group already holds a commanding position: memory. SK hynix's board last week approved 54.3 trillion won ($38.4 billion) of investment through 2031 in new semiconductor production facilities in Yongin and Cheongju. The dollar conversion is based on the won trading at about 1,414.4 per dollar Thursday. Of that, 35.2 trillion won will go toward Y2, the second of four fabs planned for SK hynix's Yongin Semiconductor Cluster. Construction is scheduled to begin in July next year, with the first cleanroom opening targeted for June 2029 to produce HBM and other advanced DRAM products. Another 19.1 trillion won will fund the M17 fab in Cheongju, where construction is scheduled to begin in February 2027 and the first cleanroom is targeted for December 2028. M17 will produce NAND flash, including products serving the rapidly expanding enterprise solid-state-drive market for AI servers. The investment illustrates how the AI infrastructure boom is spreading beyond processors. AI accelerators depend on HBM to continuously feed them data. Server DRAM provides working memory, while enterprise SSDs hold the expanding volumes of information required for training and inference. Packing those components into increasingly dense server racks, in turn, creates more heat that must be removed. SK hynix expects both DRAM and NAND demand to grow at a compound annual rate of 19 percent through 2030, citing Omdia estimates, and argues that memory is becoming part of the core infrastructure determining AI performance rather than merely another component. NAND demand in particular is expanding as AI shifts toward inference, where storage is increasingly used for techniques such as key-value caching to avoid repeatedly calculating the same information. Samsung and LG chase the heat Samsung is positioning itself at two ends of that same infrastructure chain. Its semiconductor business supplies DRAM, HBM and storage products for AI servers, while FläktGroup gives it a business supplying the systems that keep data-center equipment operating within safe temperatures. FläktGroup operates 13 production sites globally and provides systems ranging from conventional air handling and computer-room air conditioning to liquid cooling and fan-wall solutions for hyperscale and enterprise data centers. Samsung said last year that FläktGroup had worked on large hyperscale projects, including the Stargate AI infrastructure initiative, giving the Korean company access to customers already building some of the world's biggest computing facilities. The new Pune line extends that manufacturing network closer to Asian customers and will operate alongside FläktGroup's existing Noida facility in India. "The completion of this production line in Pune marks an important milestone in expanding HVAC supply beyond India to the Asia-Pacific region," said Kim Cheol-gi, executive vice president and head of Samsung Electronics' Digital Appliances business. Kim said Samsung plans to combine its AI and platform technologies with FläktGroup's HVAC capabilities to target customers including AI data centers. Samsung is not alone in chasing the heat. LG Electronics has made AI data-center cooling a strategic growth business, building what it calls a "chip-to-chiller" portfolio that stretches from cold plates attached close to processors to coolant distribution units and large chillers that remove heat from the facility. In July, LG's 600-kilowatt coolant distribution unit received Nvidia AI-infrastructure validation after meeting more than 100 technical requirements, making LG a validated supplier for Nvidia-based AI data-center infrastructure. The unit uses direct-to-chip liquid cooling, circulating coolant close to heat-generating components including GPUs and CPUs while working alongside air-cooling systems. LG plans to extend the technology into larger systems as server-rack power density continues to increase. The appeal of such systems is becoming clearer as processors grow hotter. TrendForce says thermal design power has risen from about 700 watts for Nvidia's H100 and H200 processors to more than 1,000 watts for newer generations, helping drive wider adoption of liquid cooling. The challenge is not simply preventing a server from overheating. Higher power density determines how many processors can be deployed in a rack, how closely those racks can be packed and ultimately how much computing output operators can extract from a data-center site constrained by electricity and cooling capacity. Moon cautioned against treating the AI data center as an entirely new type of infrastructure. Data centers have always underpinned computing, he said. What AI has changed is the scale of the workload and the concentration of processing power, placing much greater demands on the infrastructure surrounding the servers. Korean companies have so far been measured largely by their chips in the global AI race. Samsung, SK and LG are now trying to expand that role into storing, housing and cooling the computing power those chips enable. For Korea, the bigger opportunity may be not just to supply the memory inside the AI machine, but to capture a larger share of everything required to keep it running. AJP Takeaways • South Korean technology groups are moving beyond AI memory into the physical infrastructure surrounding it, targeting data centers, storage and cooling as global AI spending spreads across the supply chain. • Samsung's FläktGroup has opened a Pune production line capable of making up to 6,500 HVAC units annually, extending Samsung's AI exposure from memory chips to the systems needed to keep data centers cool. • SK is linking massive memory investment with plans for as much as 15 GW of AI data-center capacity, while SK hynix is spending 54.3 trillion won on new DRAM and NAND fabs in Yongin and Cheongju. 2026-08-13 16:47:27 -
HBM isn't enough: Chipmakers race to break AI's memory wall SEOUL, August 12 (AJP) - The race to build faster artificial intelligence memory is beginning to move beyond simply stacking more DRAM, as chipmakers confront a more fundamental problem: even high-bandwidth memory may not be enough to keep increasingly powerful AI processors fed with data. For the past several years, competition among memory makers has centered on successive generations of HBM, from HBM3E to HBM4 and HBM4E. SK hynix, Samsung Electronics and Micron have raced to increase bandwidth and capacity while cutting power consumption as demand from AI chipmakers such as Nvidia has surged. That demand is still accelerating. Micron Technology estimates the global HBM market will grow from about $35 billion in 2025 to around $100 billion by 2028, representing compound annual growth of roughly 40 percent. The company now expects the market to reach the $100 billion milestone two years earlier than previously forecast. But as AI models grow larger and inference workloads become more demanding, the industry's focus is broadening from the performance of individual memory chips to how memory, processors, packaging and interconnects are designed together. At the heart of the shift is the so-called "memory wall" — the widening gap between the speed at which processors can perform calculations and the speed at which data can be supplied to them. HBM has substantially eased that bottleneck by placing stacks of DRAM close to processors and allowing far larger volumes of data to move in parallel. It has not eliminated it. Higher stacks and faster bandwidth bring their own constraints, including heat, power consumption, manufacturing complexity and cost. HBM also consumes a disproportionately large share of DRAM production capacity. TrendForce estimates HBM will account for about 22 percent of total DRAM wafer input among the world's three major memory suppliers by the end of 2026 while representing only about 9 percent of total DRAM bit supply. By the end of 2027, those shares are projected to rise to 30 percent and 13 percent, respectively, as larger die sizes and growing demand put greater pressure on production capacity. The numbers expose a central trade-off. HBM provides enormous bandwidth, but producing it consumes wafer capacity far faster than it expands the industry's overall supply of memory bits. That makes simply stacking more DRAM an increasingly expensive answer to AI's data problem. As workloads expand, the challenge is shifting toward how efficiently data can move across an entire computing system — from storage and memory to processors and accelerators — without power consumption, heat or latency overwhelming performance. Still, industry experts caution against reading the proliferation of new memory concepts as evidence that HBM is about to be displaced. "There are always many concepts for new forms of memory, but it is extremely difficult for them to go through verification and commercialization and ultimately reach mass production," said Lee Jong-hwan, a professor of system semiconductor engineering at Sangmyung University. "Only a very small number actually make it all the way to mass production." That is pushing semiconductor companies to explore architectures that extend beyond conventional HBM. Intel and SoftBank subsidiary SAIMEMORY, for example, are developing Z-Angle Memory, or ZAM, a next-generation stacked DRAM architecture aimed at delivering greater capacity and lower power consumption than current HBM designs. The project is part of a broader industry effort to rethink the physical relationship between processors and memory rather than relying solely on successive upgrades to existing HBM standards. Nvidia is also approaching memory increasingly as part of a broader computing architecture. Its latest Rubin GPU integrates up to 288 gigabytes of HBM4 and delivers up to 22 terabytes per second of memory bandwidth, about 2.8 times that of its Blackwell generation. But the architecture does not rely on HBM alone. Nvidia is combining faster memory with new memory controllers, tighter compute-memory integration and technologies designed to improve data movement across the system. The message is increasingly clear: raw bandwidth still matters, but so does everything between the memory cell and the processor. The shift is unfolding as AI reshapes the wider memory industry. Counterpoint Research expects server products to account for 56 percent of global memory revenue in 2026, up sharply from 37 percent in 2025, as spending on AI infrastructure shifts demand toward high-performance memory used in data centers. For Korean memory makers, however, the change does not mean HBM is approaching the end of its growth cycle. Demand remains strong as Nvidia, AMD and other chipmakers roll out more powerful AI accelerators, while HBM4 and its successors are expected to remain central to AI infrastructure for years. Rather, the shift suggests that leadership in the next phase of AI memory will require more than producing the fastest or highest-capacity DRAM stack. SK hynix, currently at the forefront of the HBM market, has already begun positioning itself for that transition. The company has outlined plans to become what it calls a "full-stack AI memory creator," expanding beyond HBM into AI-focused DRAM and NAND products while working more closely with customers on chip architecture, workloads, power requirements and thermal design. Its "Memory as a Service," or MaaS, strategy similarly seeks to tailor memory solutions to individual AI systems rather than treating memory as a standardized component supplied after processors have already been designed. Behind the corporate terminology is a deeper change in the industry's division of labor. Memory companies historically focused on developing chips to customer specifications and manufacturing them at scale. As the memory wall becomes a larger constraint on AI performance, accelerator designers increasingly need memory, packaging and interconnect technologies optimized alongside the processor itself. That is pulling memory companies earlier into the system-design process. Even SK hynix acknowledges that HBM alone is unlikely to be the industry's final answer. At the RAISE Summit in Paris last month, Kim Ho-sik, vice president responsible for memory system research at SK hynix, said HBM was currently the best available memory technology for alleviating the memory wall but was not the ultimate solution. His remarks followed criticism from former Intel CEO Pat Gelsinger, who pointed to HBM's thermal and efficiency limitations during a panel discussion and argued that the industry would eventually need a new memory architecture to unlock AI's full potential. Kim said SK hynix was also studying other approaches to address the bottleneck while stressing that HBM remains the best technology currently available. The exchange captured the industry's emerging consensus: HBM remains indispensable, but the search for what comes alongside — and eventually beyond — it has already begun. Lee said, however, that HBM's position is unlikely to be seriously challenged in the near term, pointing to long-term supply agreements between memory makers and their major customers as an important indicator of future demand. "HBM is likely to remain firmly in place for the time being," Lee said. "The fact that long-term contracts are already in place gives memory makers visibility into how much they will need to produce several years from now and allows them to plan their investments accordingly." For Samsung Electronics and SK hynix, that could widen the battlefield considerably. Their competitive advantage was built on the ability to manufacture increasingly dense and sophisticated memory at enormous scale. The AI era is pulling them further upstream, into decisions once dominated by processor and system designers: where memory sits, how it connects and how data flows through the machine. The next memory war may therefore be decided not by who can stack the most DRAM dies, but by who helps design the AI computer around them. 2026-08-12 17:35:23 -
Hot Stock: LG Electronics fly on hopes for deeper Nvidia ties SEOUL, August 12 (AJP) - Shares of LG Electronics jumped nearly 13 percent on Wednesday, fueled by expectations that an anticipated meeting between LG Group Chairman Koo Kwang-mo and Nvidia CEO Jensen Huang could lead to deeper cooperation in robotics and artificial intelligence infrastructure. LG Electronics closed at 205,000 won ($148), up 12.82 percent from the previous session, after climbing as high as 212,500 won during trading. Koo is visiting Nvidia's headquarters in Silicon Valley this week for talks with Huang, according to industry sources, drawing investor attention to potential expansion of ties between the two companies. The two executives last met in Seoul in June, when they discussed potential cooperation in areas including robotics, AI infrastructure and autonomous driving. Investor expectations have also been supported by growing interest in so-called physical AI, which extends artificial intelligence into machines and real-world systems such as robots. Morgan Stanley recently identified LG Electronics as a potential beneficiary of the trend, pointing to its businesses spanning AI-powered homes, robotics, commercial heating, ventilation and air conditioning systems and cooling solutions for AI data centers. The broader LG Group also has businesses that could feed into physical AI infrastructure, including sensing and optical components from LG Innotek, batteries from LG Energy Solution and smart factory capabilities from LG CNS. LG Electronics has meanwhile been expanding beyond its traditional consumer appliance business. On Wednesday, the company launched a new diagnostic monitor cleared by the U.S. Food and Drug Administration as part of an expansion of its B2B medical display business. Wednesday's rally followed a sharp pullback in the stock, which fell 4.37 percent in the previous session. The shares had surged to around 390,000 won in early June amid expectations surrounding Koo and Huang's previous meeting before retreating sharply in subsequent weeks. 2026-08-12 15:43:07 -
Money may buy happiness: SK hynix happier than Samsung SEOUL, August 12 (AJP) - Who says money can't buy happiness? SK hynix ranked among the top 3 percent of South Korean companies for employee happiness while archrival Samsung Electronics placed around the top 60 percent, according to a workplace survey released Wednesday, underscoring how the AI memory boom is translating into starkly different perceptions of reward inside Korea's two biggest chipmakers. The findings came from the latest Blind Index, compiled by workplace platform Blind using a methodology jointly developed with the Korea Labor Institute. The survey covered 34,372 South Korean employees between July and December last year and does not reflect this year's earnings and bonus windfall. The divergence is notable as the two chipmakers compete for talent as fiercely as they do for leadership in artificial intelligence memory. Money may be part of the answer. SK hynix allocates 10 percent of annual operating profit to its profit-sharing, or PS, program after removing a previous payout cap. Based on this year's projected earnings, employees could receive around 700 million won ($486,000) each on average, although actual payouts vary by rank and individual performance. Samsung Electronics has moved closer to that model this year after labor and management agreed to introduce a separate bonus for its Device Solutions chip division on top of its existing Overall Performance Incentive, or OPI, scheme. Samsung has traditionally calculated OPI based on economic value added, with payouts capped at 50 percent of annual salary. The new DS bonus instead draws from 10.5 percent of agreed business performance and carries no payout ceiling. Under current profit assumptions, employees in Samsung's highly profitable memory business could receive close to 600 million won in combined incentives, while payouts elsewhere could be far smaller depending on business performance. The difference is not only how much employees receive, but how they receive it. SK hynix pays 80 percent of its PS award in cash in the year it is earned, with the remaining 20 percent deferred over the following two years. Employees can also elect to receive part of their bonus in company shares. Samsung's new semiconductor bonus, by contrast, will be paid entirely in Samsung Electronics shares after tax. One-third can be sold immediately, another third after one year and the remainder after two years. The scheme is set to run through 2035 and is subject to annual operating-profit thresholds, marking Samsung's attempt to tie semiconductor compensation more directly to the profits generated by its chip business. Public reviews on Blind also point to a gap between the companies. SK hynix currently carries an overall employee rating of 3.0 out of 5, compared with Samsung Electronics' 2.8. SK hynix scores 3.3 for compensation and benefits and 2.9 for corporate culture, while Samsung scores 3.0 and 2.7, respectively. One person familiar with Samsung's workplace culture attributed part of the lower employee sentiment to what the source described as an intensely performance-driven environment, which has also exposed widening disparities between highly profitable chip operations and weaker businesses during the semiconductor boom. "Samsung Electronics is highly goal-oriented, with individual performance often prioritized over collaboration," the source told AJP. "There are also internal systems through which employees can anonymously report their colleagues, which can create an atmosphere where people find it difficult to fully trust those they work with." Blind's publicly available Samsung Index points to broader concerns over workplace culture. Samsung ranks relatively low on measures including employees' sense of meaning and importance in their work, workplace bonds, work-life balance and psychological safety, according to the platform. The gap does not mean sentiment at SK hynix is uniformly positive. Recent verified employee reviews on Blind praise compensation and benefits but also cite heavy workloads and dissatisfaction with management and labor negotiations, suggesting even some of the industry's richest bonuses have limits in determining workplace happiness. Across corporate South Korea, employee sentiment deteriorated more broadly. Only 33 percent of companies surveyed received happiness scores of 50 or higher, down from 47 percent a year earlier, according to Blind. The deterioration has not translated into greater employee mobility. The proportion of workers who attempted to change jobs over the past year fell despite lower happiness scores, suggesting a weaker labor market is keeping more dissatisfied workers where they are. SK hynix ranked among the top 3 percent of South Korean companies for employee happiness while archrival Samsung Electronics placed around the top 60 percent, according to a workplace survey released Wednesday, reflecting how generous reward in cash instead of stock option influenced employees in rating their contentment with their employers. The findings came from the latest Blind Index, compiled by workplace platform Blind using a methodology jointly developed with the Korea Labor Institute. The survey covered 34,372 South Korean employees between July and December last year. The divergence is notable as the two chipmakers compete for talent and leadership in artificial intelligence memory. Public reviews on Blind also point to a modest gap between the companies. SK hynix currently carries an overall employee rating of 3.0 out of 5, compared with Samsung Electronics' 2.8. SK hynix scores 3.3 for compensation and benefits and 2.9 for corporate culture, while Samsung scores 3.0 and 2.7, respectively. One person familiar with Samsung's workplace culture attributed part of the lower employee sentiment to what the source described as an intensely performance-driven environment that resulted in a yawning chasm between profit-making chip and non-chip operations during the chip boom. "Samsung Electronics is highly goal-oriented, with individual performance often prioritized over collaboration," the source told AJP. "There are also internal systems through which employees can anonymously report their colleagues, which can create an atmosphere where people find it difficult to fully trust those they work with." Blind's publicly available Samsung Index offers some support for broader concerns over workplace culture. Samsung ranks relatively low on measures including the perceived meaning and importance of employees' work, workplace bonds, work-life balance and psychological safety, according to the platform. The gap comes despite mixed sentiment at SK hynix itself. Recent verified employee reviews on Blind praise compensation and benefits, while others cite heavy workloads and dissatisfaction with management and labor negotiations, suggesting the survey ranking does not translate into uniformly positive employee sentiment.' SK hynix employees are moving to form a new unified union spanning production and technical-office workers as they seek greater bargaining power in a dispute over management's proposed changes to the bonus system. More than 3,500 workers joined the preparatory group within three days of its launch, equivalent to about 10 percent of the company's workforce. Across corporate South Korea, workplace happiness deteriorated more broadly. Only 33 percent of companies surveyed received happiness scores of 50 or higher, down from 47 percent a year earlier, according to Blind. The decline, however, has not translated into greater employee mobility. The proportion of workers who attempted to change jobs over the past year fell despite lower happiness scores, suggesting a weaker labor market may be keeping dissatisfied employees in place. 2026-08-12 13:58:45 -
SK hynix workers move to unify union to up bargaining power SEOUL, August 12 (AJP) - Workers at SK hynix are moving to establish a new union spanning production and office employees, with organizers seeking majority representation to strengthen bargaining power at the world's leading high-bandwidth memory maker. The proposed SK hynix Integrated Labor Union had attracted more than 1,250 members as of Wednesday after its organizing committee filed for official registration on Aug. 8, according to industry sources. The union could formally launch as early as Wednesday. Organizers aim to recruit about 18,000 members, more than half of SK hynix's roughly 35,000 employees, in an effort to secure majority-union status. SK hynix currently operates under a multiple-union structure, with three existing unions separately representing production workers in Icheon and Cheongju and technical and office employees. Organizers of the new union argue that the fragmented structure has weakened employees' bargaining leverage with management. The new group plans to remain independent of South Korea's two major umbrella labor organizations and give individual members equal voting rights on major decisions. It also pledged greater disclosure of negotiations and restrictions on political activities by union leadership. Profit-sharing is expected to be a key issue. Organizers said they would seek to preserve cash payments under SK hynix's profit-sharing, or PS, scheme and oppose attempts to replace them with shares if such changes require collective employee consent under labor law. The union has also sought advice from a cross-company union at Samsung Electronics and is considering future cooperation as workers respond to potential changes in profit-linked compensation schemes. Its immediate influence, however, could be limited. SK hynix is already conducting this year's wage and collective bargaining negotiations with its three existing unions, making it uncertain whether the new organization will be able to formally join the current round of talks. Organizers said that if direct participation is not possible, they plan to seek observer access to negotiations, disclosure of bargaining developments and other channels to relay employee views. 2026-08-12 10:53:45