Retail News CRM

Tag: tsmc

  • TSMC Explores Rare-Earth Alternatives for Factory Machinery

    TSMC Explores Rare-Earth Alternatives for Factory Machinery

    TSMC is exploring alternatives to rare-earth components across its factory equipment to curb risks that Beijing’s export controls could disrupt plants manufacturing the world’s most advanced chips, three people familiar with the matter said.

    The world’s largest contract chipmaker surveyed automation suppliers on using rare-earth-free motors after China placed certain rare-earth materials, magnets and processing technology under export licensing rules last year.

    A shortage of specialised magnets in factory motors threatens chip output for AI systems, smartphones and data centres, though talks remain early and TSMC declined to comment.

    Supply Checks on Factory Floor Motors

    Specialised magnets made with rare earths drive precision motors in automated wafer-handling systems and cleanroom equipment. Without these components, high-throughput manufacturing lines ground to a halt, cutting off output bound for smartphone makers, consumer tech brands and data centre operators. TSMC has approached equipment suppliers to gauge whether alternative motor designs can match current production tolerances without relying on Chinese materials.

    Talks remain at an early phase and have not triggered wide equipment replacements across operational plants. TSMC declined to comment on vendor communications.

    Vulnerabilities in Advanced Electronics Pipelines

    Hardware brands across Asia track fabrication inputs closely because component disruptions cascade rapidly into consumer hardware shortages. A bottleneck in factory machinery slows capacity additions for cutting-edge nodes, lifting wafer costs and tightening component supply for consumer electronics brands already balancing thin margins.

    For equipment vendors, redesigning factory tools without rare-earth magnets requires heavy engineering adjustments. Permanent magnets provide the highest torque density for compact industrial robotic arms, meaning substitutions often demand larger machinery footprints or reduced operating speeds. Capital expenditure across upcoming wafer fabrication projects in Taiwan and the United States will reflect the cost of these redesigns.

    Government Initiatives Back Mineral Independence

    Export restrictions imposed by Beijing last year placed several rare-earth materials, industrial magnets and associated extraction tech under strict licensing regimes. The regulatory clampdown prompted tech-heavy economies across the Pacific to fund alternative supply chains and local extraction capacity.

    In June, American Institute in Taiwan director Raymond F. Greene confirmed that Washington is coordinating with Taipei to secure critical minerals and build domestic processing capacity. Domestic efforts are also expanding inside Taiwan. In February, Economy Minister Kung Ming-hsin said the government would support the Industrial Technology Research Institute to establish a pilot production facility within three years to meet an estimated 50 percent of domestic rare-earth demand.

    The state-backed research institute confirmed it is working with the Ministry of Economic Affairs on an execution plan for the mineral initiative.

  • Rapidus Partners with 17 Design Firms to Secure 2Nm Chip Buyers

    Rapidus Partners with 17 Design Firms to Secure 2Nm Chip Buyers

    Japan’s state-backed chipmaker Rapidus has partnered with 17 design firms to secure buyers for its planned two-nanometre foundry. The venture holds 15 billion dollars in government funding.

    Executives must book production orders before commercial fabrication starts in the second half of the next fiscal year.

    Partners include American design software provider Synopsys and Indian technology services group Infosys. These alliances allow Rapidus to offer external customers design support alongside wafer fabrication. Early software integration resolves bottlenecks before physical manufacturing starts in Chitose, Hokkaido. Rapidus is developing its two-nanometre process with IBM and has already begun pilot line operations.

    Chasing Market Share in Advanced Silicon

    Landing customer commitments remains the primary operational hurdle. Taiwan Semiconductor Manufacturing Co dominates cutting-edge contract manufacturing. South Korea’s Samsung Electronics started producing two-nanometre chips last year, and Intel continues its own foundry expansion. Rapidus is pitching its capacity to consumer electronics and enterprise hardware designers shut out of production queues at TSMC.

    Switching supplier allocations to an untested foundry carries heavy risks for device makers. Global tech brands operate on strict product refresh cycles. Fabrication delays or yield faults quickly derail device launches across retail channels. Hardware brands want supply chain alternatives to hedge geopolitical tension around the Taiwan Strait. Still, migrating proprietary chip designs away from established foundries carries significant financial exposure.

    Rebuilding Japan’s Semiconductor Base

    National market share in global semiconductor manufacturing has fallen from roughly 50 per cent in the 1980s to below 10 per cent today. Prime Minister Sanae Takaichi is relying on Rapidus to anchor a broader industrial revival. That effort links directly to domestic artificial intelligence initiatives like the Noetra robotics project.

    Local suppliers of silicon wafers, precision chemicals and lithography equipment also gain direct demand if Rapidus succeeds. Commercial production in Hokkaido will shorten procurement cycles for regional device makers. Those firms currently export raw materials to Taiwan and South Korea for fabrication before importing finished silicon packages.

    Foundry Economics and Operational Challenges

    Fab viability depends on nonstop machine utilisation and high production yields. Even well-funded incumbents face margin compression when wafer defect rates spike during early-stage node transitions.

    Running a chip fab around the clock, maintaining stable, consistent yields and continuing production in a way that makes the business viable is extremely difficult

    Said Kazuyoshi Saito, an analyst at IwaiCosmo Securities.

    Capital recovery will take years before operations turn cash-positive. Rapidus management has set an initial public offering target for around the fiscal year ending March 2032 to open private equity participation and reduce reliance on state funding.

    Milestones Ahead of Commercial Ramp

    Engineering teams must now convert preliminary design consultations with the 17 partners into signed volume supply contracts. Pilot yields from the Hokkaido facility will face customer evaluation throughout 2027 before volume tool installations begin for commercial wafer output.

  • Taiwan Banks Ration Credit After US$126 Billion Tech Borrowing Spree

    Taiwan Banks Ration Credit After US$126 Billion Tech Borrowing Spree

    Taiwanese banks extended a record NT$4 trillion, or US$126 billion, in loans during the first seven months of the year. The borrowing surge triggered credit rationing across the island.

    Demand for working capital from suppliers serving Nvidia Corp and Microsoft Corp pushed monthly bank lending to an all-time high of NT$1.04 trillion in July. That upended years of low interest rates.

    Local lenders raised corporate borrowing rates between 50 and 90 basis points. Margins for top-tier syndicated corporate facilities climbed 10 to 20 basis points. Mortgage rates reached almost 3 percent across the island, up from 2.2 percent a year earlier. Tech manufacturers need quick cash to secure semiconductor components, build inventory, and expand production lines for orders linked to Taiwan Semiconductor Manufacturing Co and global artificial intelligence infrastructure programs. Many suppliers now demand loan approvals within two months, cutting a third off the standard 90-day evaluation window.

    Surging Working Capital and Squeezed Margins

    Working capital loans to businesses and individual borrowers rose by NT$823 billion in July alone. That expanded nearly eight times faster than residential mortgage growth. The spike pushed the banking system loan-to-deposit ratio to 72.96 percent, near its historical ceiling of 73.07 percent recorded in 2020. Average liquidity coverage ratios fell to 115.52 percent in July. While still above the statutory 100 percent threshold, it was the sharpest four-month liquidity drop in four years.

    Commercial banks are shifting balance sheets toward higher-yielding corporate facilities and short-term consumer credit lines. Unsecured personal loans yield fatter spreads than residential mortgages, which face central bank cooling measures and property transaction curbs. At the same time, retail depositors accelerated withdrawals to buy listed technology shares. That pulled deposits from regional institutions and tightened wholesale funding conditions.

    Credit Shut Out for Builders and Exporters

    Capital reallocation is starving non-technology businesses of routine working credit. Commercial property developers and homebuilders have seen credit pipelines collapse. At one regional lender, weekly approvals for land acquisition and construction financing dropped from 30 deals to zero. Loan officers instructed developers to halt new submissions entirely.

    Traditional industrial exporters, already dealing with tariff headwinds and sluggish consumer goods demand abroad, face tighter credit terms as bank committees favor electronics manufacturing. Taiwan’s economy expanded 13 percent year on year in the second quarter. The expansion is lopsided. Industrial estates housing advanced packaging and chip fabrication run around the clock, while consumer-facing businesses face rising financing hurdles and slowing real estate sales.

    Offshore Syndication Pullback and Regional Spillover

    Liquidity strains are shifting credit dynamics across Southeast Asia, where Taiwanese commercial banks have long served as anchor lenders for syndicated facilities. Taiwanese institutions are scaling back commitments to offshore transactions across Singapore, Hong Kong, and Vietnam. They are pricing themselves out of lower-margin deals or declining syndication tranches to preserve balance-sheet capacity for domestic tech clients. For regional borrowers who relied on Taipei banks for tight margins, financing costs will rise as Japanese and European lenders step in with higher pricing hurdles.

    Structural risk sits directly on balance sheets overconcentrated in single-sector capital expenditure. Banks are directing most of their incremental loan books into hardware suppliers tied to volatile data-center construction schedules. Any pause in hyperscaler procurement cycles will hit banking asset quality swiftly. Traditional suppliers, service firms, and real estate operators unable to access credit today face balance-sheet stress that could outlast the tech cycle.

    Central Bank Coordination and Regulatory Scrutiny

    The credit squeeze follows months of aggressive capital deployment across the hardware supply chain. Micron Technology accelerated its memory production investments in Taiwan to NT$1.6 trillion by mid-year, concentrating high-bandwidth capacity on the island. Total outstanding residential mortgages still climbed 4.6 percent to reach NT$11.95 trillion in August. Bank appetite for long-term mortgage debt has withered in favor of revolving corporate credit.

    Central bank Governor Yang Chin-long has initiated talks with state-backed lenders, commercial institutions, and Chunghwa Post Co, the postal savings system holding the island’s largest pool of retail deposits. The meetings aim to coordinate liquidity relief for small and medium-sized enterprises. The Financial Supervisory Commission confirmed deposit growth currently tracks overall loan creation without immediate systemic failures, but said it will step up surveillance of bank capital allocations.

    Officials from the central bank will hold follow-up consultations with commercial bank treasurers to review quarterly liquidity ratios and evaluate whether non-technology corporate borrowers face structural lending barriers through the final quarter.

  • TSMC Affiliate Vanguard Eyes Second Singapore Plant as First Sells Out

    TSMC Affiliate Vanguard Eyes Second Singapore Plant as First Sells Out

    Taiwan Semiconductor Manufacturing Co. Affiliate Vanguard International Semiconductor inaugurated its first advanced chipmaking plant in Singapore on Monday, with capacity already sold out amid an artificial intelligence infrastructure boom.

    The plant was constructed in less than two years as chipmakers race to meet demand for data center components.

    With the initial site opening completely full, the company is already eyeing expansion and planning a second facility.

    Surging Data Center Component Intake

    Surging demand for AI computing has outstripped regional supply chains. Packaging and wafer fabrication lines filled up before commercial operations even began. Enterprise data center operators are securing long-term component capacity years ahead to avoid hardware bottlenecks.

    For technology equipment suppliers across Southeast Asia, the quick capacity fill guarantees immediate local component availability for server assembly and power management hardware. Device manufacturers that delayed long-term supply agreements now face extended lead times. Available volume is shifting to hyperscalers.

    Singapore Role in Asian Semiconductor Shift

    Foundries are establishing high-specification manufacturing clusters outside Taiwan and mainland China. Southeast Asia has drawn substantial capital expenditure from global semiconductor firms seeking neutral production hubs with established infrastructure and logistics links.

    This investment pairs NXP automotive and industrial silicon expertise with Vanguard manufacturing discipline. The combination insulates downstream consumer electronics and automotive assemblers across the Asia-Pacific region from single-location supply disruptions.

    Fast-Track Construction and Factory Timelines

    Vanguard and NXP broke ground on the initial Singapore site to capture expanding demand for specialty and power management semiconductors. The venture finished construction within twenty-four months to meet rapid AI hardware deployment schedules.

    Earlier expansion programs across the region faced construction delays and tool installation bottlenecks. By contrast, the Singapore project maintained its delivery timetable through streamlined municipal permitting and fast-tracked site preparation.

    Next Steps for the Second Facility

    Planning for the second Singapore plant now hinges on equipment procurement schedules and utility allocations from local authorities. VisionPower must finalize technical specifications for the proposed advanced production lines before committing total capital expenditure numbers.

    Directors on the joint venture board will next review the project to decide on the final construction timeline and the exact process geometry for the second facility.

  • TSMC to Build Five More Packaging Plants at Chiayi Science Park

    TSMC to Build Five More Packaging Plants at Chiayi Science Park

    Taiwan Semiconductor Manufacturing Co plans to build five additional advanced packaging plants at Chiayi Science Park in southwestern Taiwan, following the start of mass production at Phase I in June.

    The project will bring TSMC’s total advanced packaging facilities at the Chiayi site to 10, creating the world’s largest packaging hub for high-performance semiconductor components.

    Taiwan’s National Science and Technology Council has opened a one-month public consultation for Phase II expansion at the park in Taibao City, which local industries view as the launchpad for Phase III development. The planned Phase III zone covers 198.65 hectares, primarily leased from Taiwan Sugar Corp, to support technologies related to artificial intelligence, heterogeneous packaging, silicon photonics, high-speed broadband networks, quantum technology and net zero emissions.

    Scaling Up Advanced Packaging Capacity

    TSMC already operates two advanced packaging lines in Phase I of the Chiayi park, where mass production began in June. Equipment installation is now underway in Phase II. The company intends to complete three packaging facilities in Phase II by 2031, while preliminary site preparation for Phase III has begun.

    Building out Chiayi links the location to existing science parks in Tainan, Kaohsiung and Pingtung County to the south, as well as hubs in Taichung and Hsinchu to the north. Together, the interconnected parks form an integrated manufacturing corridor across western Taiwan designed to keep advanced fabrication close to domestic engineering teams.

    CoWoS Demand From Global Chip Buyers

    Surging orders for chip-on-wafer-on-substrate packaging from Nvidia, Advanced Micro Devices and Apple serve as the primary commercial engine behind the Chiayi project. Advanced packaging binds multiple silicon dies onto a single substrate, giving AI accelerators and mobile processors the memory bandwidth required for high-density computing workloads without relying solely on smaller nanometer lithography.

    For hardware manufacturers and consumer electronics brands, packaging bottlenecks have dictated device delivery schedules for more than two years. Concentrating 10 packaging plants in Chiayi addresses that supply constraint directly, giving key enterprise and consumer hardware clients guaranteed assembly allocations through the end of the decade. The concentration of advanced packaging on the island also raises the operational stakes for global electronics supply chains, which remain heavily dependent on Taiwan’s local power grid, water supplies and regional transport networks.

    Taiwan Infrastructure and Regional Competition

    While TSMC expands fabrication operations overseas in the United States, Japan and Germany, its most advanced packaging processes remain anchored in Taiwan. Packaging clusters require tight proximity to the leading-edge wafer fabs that produce raw silicon slices, making domestic expansion far more cost-effective than duplicating complex multi-die assembly networks abroad.

    Competitors such as Intel and Samsung continue to build out proprietary advanced packaging services to win third-party foundry clients. TSMC’s accelerated buildout in Chiayi prevents rivals from capturing unmet packaging demand from top-tier chip designers who cannot secure sufficient domestic capacity elsewhere.

    What Comes Next in the Planning Process

    Capital spending supports the speed of the expansion. In July, TSMC lifted its full-year capital expenditure forecast to a band between US$60 billion and US$64 billion, up from an earlier estimate of US$52 billion to US$56 billion announced in April. The Ministry of Economic Affairs reports that TSMC operates 19 advanced wafer fabs and packaging plants across Taiwan, with 13 additional facilities under construction in Hsinchu, Taichung, Chiayi, Tainan and Kaohsiung.

    Following the conclusion of the public comment window for the Chiayi expansion, the National Science and Technology Council will complete a formal feasibility study for submission to the Executive Yuan. Once the central government approves the plan, the project will advance to statutory urban planning reviews and environmental impact assessments before Phase III construction contracts are awarded.

  • Taiwan Targets Silicon Photonics to Cut AI Chip Power Use by 50 Percent

    Taiwan Targets Silicon Photonics to Cut AI Chip Power Use by 50 Percent

    Taiwan is mobilising its chipmaking sector to commercialise silicon photonics and copackaged optics, with the Ministry of Economic Affairs backing efforts to overcome artificial intelligence computing bottlenecks.

    The island holds over 90 percent of global manufacturing capacity for chips at 7 nanometers or below, while more than 30 firms including Taiwan Semiconductor Manufacturing Co and MediaTek Inc formed an industry alliance in 2024.

    Silicon photonics replaces conventional copper wiring between chips with optical signals. Because optical transmission generates negligible heat, the technology cuts device power consumption by 30 to 50 percent, according to the Ministry of Economic Affairs. That saving eases cooling limits in high-density AI data centres. It also clears the 1.6-terabit-per-second bandwidth ceiling that hampers electronic links.

    Optics Replace Copper Interconnects

    Heavy capital is pouring across the hardware supply chain. Nvidia invested US$4 billion into silicon photonics development in March. The chipmaker needs architectures that sustain real-time AI workloads and high-definition streaming without overheating server racks.

    Taiwanese authorities have folded the technology into their New 10 Major AI Infrastructure Projects. Funding flows through the government’s A+ Enterprise Innovation research programme. The scheme subsidises domestic research teams and equipment developers building local manufacturing tooling.

    Alliance Mobilises Heavyweights

    Execution on the ground rests on the Silicon Photonics Industry Alliance, a consortium established in 2024 with industry group SEMI. The group brings together more than 30 technology suppliers. Members include Taiwan Semiconductor Manufacturing Co, ASE Technology Holding, MediaTek and Hon Hai Precision Industry.

    Asian hardware vendors and server assemblers face shifting procurement cycles. Traditional printed circuit board layouts will yield to integrated optical packaging. As a result, component suppliers must retool production lines for optical transceivers and precision glass substrates.

    Manufacturing risks centre on packaging yields and costs. Integrating laser sources and optical waveguides directly alongside silicon dies requires packaging tolerances tighter than standard wire bonding. These yield hurdles could delay volume delivery.

    Race for Next-Generation Packaging

    Foundry and packaging operators have spent years researching optics to counter the slowdown of traditional transistor scaling. Physical node shrinking now delivers diminishing returns. Advanced packaging formats like copackaged optics have become the primary path to computing efficiency.

    Prototype lines are now running across alliance members to finalise copackaged optics standards before volume production begins for 2026 data centre hardware cycles.

  • Taiwan Pitches Global Chip Alliances as Market Heads for 1.5 Trillion Dollars

    Taiwan Pitches Global Chip Alliances as Market Heads for 1.5 Trillion Dollars

    Taiwan President William Lai told tech executives in Taipei that international chip partnerships will anchor an industry projected to exceed 1.5 trillion dollars this year.

    Foreign semiconductor leaders continue to pour capital into the island, led by Nvidia’s annual procurement and investment topping NT$3 trillion ($94.84 billion). Micron Technology has committed more than NT$1.4 trillion to local operations, while Advanced Micro Devices pushed its research spending in Taiwan past NT$300 billion.

    TSMC’s Overseas Buildout

    Taiwan Semiconductor Manufacturing Co is matching domestic research with heavy spending abroad to insulate buyers against supply shocks. The world’s largest contract chipmaker announced an additional $100 billion commitment to its Arizona facilities in July, while its plant in Kumamoto, Japan, continues on schedule.

    In Europe, TSMC expects its Dresden fabrication facility to begin commercial chip production before the end of next year. That site will supply automotive and industrial customers across the European Union.

    For consumer electronics makers and device brands across Asia, the dual strategy offers reassurance. Taiwan is retaining cutting-edge wafer fabrication and packaging on home soil while duplicating mature and specialized capacity in Western markets to guarantee steady silicon delivery during regional crises.

    State Support for Next-Gen Tech

    Taipei plans to back corporate spending by funding core infrastructure, including power generation, water supplies, land access, and computing capacity. State research backing will focus on silicon photonics, quantum computing, and artificial intelligence robotics.

    US Undersecretary of State for Economic Affairs Jacob Helberg told attendees via video link that concentration without resilience creates systemic vulnerabilities. He pointed to the Pax Silica framework, an alliance designed to secure technology supply lines among trusted trade partners.

    Discussions continue this week as the Semicon Taiwan trade exhibition runs through Friday at the Taipei Nangang Exhibition Center.

  • TSMC Agrees to Produce Three 3Nm Chips for Xiaomi

    TSMC Agrees to Produce Three 3Nm Chips for Xiaomi

    Taiwan Semiconductor Manufacturing agreed to produce three custom chips for Xiaomi, including a next-generation three-nanometre smartphone processor starting in 2026.

    The contract ties the $2.0 trillion Taiwanese foundry directly into the Chinese brand’s hardware expansion across flagship handsets, smart devices and automotive platforms. Xiaomi plans to deploy the primary 3nm design, designated the Xring O3, in its premium smartphones before rolling out two companion processors for consumer artificial intelligence devices and autonomous vehicle controls.

    Expanding beyond data centres

    Adding Xiaomi diversifies TSMC’s advanced-node order book at a time when top-tier 3nm wafer allocation has remained heavily concentrated among Western computing and mobile clients. Handset manufacturers in Asia have spent three years attempting to bring proprietary silicon in-house to reduce their dependence on merchant chipmakers. Xiaomi’s commitment to custom designs manufactured on TSMC’s cutting-edge lithography mirrors earlier silicon strategies from rival hardware makers, though extending those designs into vehicle autonomy widens the operational scope.

    For consumer tech brands in Asia, controlling chip architecture allows tighter software integration across connected ecosystems, from living-room appliances to electric sedans. The arrangement secures advanced fabrication capacity for Xiaomi while providing TSMC with volume demand outside its core server and cloud computing base.

    Wafer volume targets

    Initial commercial success hinges on the manufacturing yield and delivery pace of the Xring O3 line during 2026. Market performance will depend on the speed at which Xiaomi ramps retail shipments of its 3nm handsets and incorporates the subsequent automotive silicon into its production vehicles.

  • Asian Tech Giants Taiwan and South Korea Form AI Hardware Alliance Amid Surging Demand

    Asian Tech Giants Taiwan and South Korea Form AI Hardware Alliance Amid Surging Demand

    Long-standing technological competitors Taiwan and South Korea are joining forces to address the booming global demand for AI accelerators, the specialized hardware critical for training generative AI models. This alliance marks a significant shift from decades of intense rivalry to a necessary partnership in the digital age.

    The insatiable demand from tech giants for computing power has strained the capacity of key accelerator suppliers like Nvidia and AMD. Their products rely on an increasingly interconnected supply chain linking Taiwan and South Korea, making cooperation essential to meet the massive market need.

    Nvidia CEO Jensen Huang highlighted the urgency at Computex 2026 in Taipei, personally requesting more memory wafers from South Korean firm SK Hynix. This collaboration underscores that no single country can satisfy the current scale of AI infrastructure expansion alone.

    From Intense Competition to Interdependence

    For decades, South Korea and Taiwan fiercely competed for supremacy in technologies ranging from televisions and smartphones to memory chips. Both nations, former Japanese colonies, built their economies on exports, initially focusing on labor-intensive goods before advancing to electronics. This rivalry saw Taiwanese brands like Acer and Asus compete against South Korean giants Samsung and LG.

    South Korea often pulled ahead due to the integrated structure of its conglomerates, such as Samsung, which controlled multiple stages of the supply chain. A notable example involved Samsung reportedly aiming to dominate industries where Taiwanese firms competed, though Samsung denied such a specific ‘Kill Taiwan’ strategy.

    Taiwan’s strength, however, lay in its highly specialized contract manufacturing ecosystem. Unlike South Korea’s vertically integrated companies, Taiwan developed a dense network of firms each focusing on a narrow part of the manufacturing process. This specialization allowed for greater flexibility and adaptability, enabling enduring partnerships with global tech firms like Apple, which eventually made Taiwan Semiconductor Manufacturing Company (TSMC) its primary chip supplier over Samsung.

    Taiwan’s AI Hardware Dominance and Future Outlook

    In the current AI era, Taiwan’s multifaceted ecosystem has proven highly advantageous. Taiwanese companies now hold dominant positions in crucial niches, from power supplies to chip packaging and advanced cooling systems. This allows global customers to select and combine suppliers, maintaining production flexibility. Nomura’s May report ranked Taiwan first globally in AI hardware production, with its AI-related exports contributing to a 15-year high economic growth of 8.7% last year.

    This economic divergence saw Taiwan surpass South Korea in GDP per capita in 2025 for the first time in over two decades. However, this has not deepened the rivalry; instead, the AI boom has fostered greater interdependence. South Korea excels in High Bandwidth Memory (HBM), with SK Hynix and Samsung together accounting for approximately 80% of global HBM production. These critical memory components are then integrated with computing chips from Taiwan’s TSMC using advanced packaging techniques, a final step in creating AI accelerators.

    This dynamic has resulted in South Korea achieving a record trade surplus with Taiwan last year. The relationship is best described as a ‘frenemy’ dynamic, driven by the sheer scale of AI demand. Despite lingering competitive sentiments, the complementary strengths of Taiwan’s contract chipmaking dominance and South Korea’s memory market stronghold make them indispensable partners in powering the next wave of artificial intelligence.

  • Taiwan Semiconductor Manufacturing Emerges as Clearer Investment Choice over SoundHound AI

    Taiwan Semiconductor Manufacturing Emerges as Clearer Investment Choice over SoundHound AI

    Taiwan Semiconductor Manufacturing Company (TSMC), the world’s largest contract chipmaker, has been identified as a significantly stronger investment choice compared to conversational AI firm SoundHound AI for the year 2026. This assessment, rooted in a comprehensive financial and operational comparison, positions TSMC as a robust foundation for the global technology ecosystem, including critical support for Asia’s burgeoning retail and consumer technology sectors.

    TSMC’s Dominance and Financial Strength

    TSMC’s financial performance in fiscal year 2025 demonstrated remarkable strength, with revenues soaring to approximately $120.3 billion, a 33% increase from the previous year. The company recorded a net income of about $54.3 billion, yielding a net margin of 45.1%. This profitability is bolstered by its role as a dedicated foundry, manufacturing advanced chips that power everything from high-performance computing to smartphones. As of December 2025, TSMC maintained a low debt-to-equity ratio of 0.2x and a healthy current ratio of 2.5x, indicating strong financial stability. Free cash flow for the year reached approximately $34.3 billion, underscoring its operational efficiency and ability to fund ongoing expansion. This makes TSMC a cornerstone for Asian electronics manufacturing and by extension, the retail chains dependent on these devices.

    The company’s strategic importance extends to its global manufacturing footprint, with facilities across Taiwan, China, and the United States, serving over 500 customers. Its advanced chip production is essential for the AI industry, with high-performance computing now accounting for nearly two-thirds of its total revenue. TSMC’s continuous investment in cutting-edge fabrication technologies, despite annual billions spent on new factories, is crucial for maintaining its market leadership against rivals like Intel and Samsung. For Asian markets, this ensures a reliable supply chain for next-generation consumer electronics and enterprise solutions.

    SoundHound AI’s Growth Amidst Challenges

    In contrast, SoundHound AI, while showing rapid growth, faces a more challenging path. The company, which provides specialized voice software for sectors like automotive, retail, and hospitality, reported revenues of nearly $168.9 million in FY 2025, a growth rate of 99.4%. However, this growth came with a net loss of approximately $14.0 million, resulting in a negative 8.3% net margin. The company’s strategy involves aggressive growth through acquisitions, such as LivePerson and Amelia, which can introduce integration complexities and higher costs. Its balance sheet as of December 2025 showed a debt-to-equity ratio of 0.0x and a current ratio of 4.6x, but free cash flow remained negative at $103.1 million.

    SoundHound AI operates in a highly competitive landscape against larger technology firms like Microsoft and Alphabet, which possess significant resources. The company has also contended with internal control weaknesses and ongoing legal challenges. While its agentic AI software finds traction with partners like Casey’s convenience stores and MUSC Health, its financial scale and profitability remain far behind TSMC. For retail and hospitality businesses in Asia considering voice AI solutions, the long-term stability and competitive resilience of providers like SoundHound AI become key considerations.

    RetailNews Asia notes that while the allure of high-growth tech firms like SoundHound AI can be strong, the foundational importance and robust financial health of companies like TSMC offer a more predictable, albeit less explosive, investment outlook for those backing the region’s vast consumer tech ecosystem. Similar to how other regional manufacturing giants provide stability, TSMC’s role is critical for the continuous innovation seen across Asian retail and technology.

  • TSMC’s AI Chip Capacity Challenges Offer Samsung A Pricing Advantage

    TSMC’s AI Chip Capacity Challenges Offer Samsung A Pricing Advantage

    Taiwan Semiconductor Manufacturing Company (TSMC), a global leader in chip manufacturing, is reportedly struggling with production constraints for its advanced AI chips. These bottlenecks are primarily linked to the intricate Chip-on-Wafer-on-Substrate (CoWoS) packaging technology, which is critical for high-performance computing components used in artificial intelligence.

    This production challenge at TSMC presents a strategic advantage for its main competitor, Samsung. With demand for AI chips surging and TSMC’s capacity limited, Samsung finds itself in a stronger position to negotiate higher prices for its competing memory products, particularly High Bandwidth Memory (HBM). HBM is a vital component often bundled with AI chips.

    Market Dynamics and Pricing Power

    The current situation highlights the intense competition and intricate supply chain dynamics within the semiconductor industry. As AI development accelerates, the ability to produce these complex chips and their associated components at scale becomes a major determinant of market leadership. TSMC’s temporary hurdles with CoWoS packaging could give Samsung a window to capture a larger share of the lucrative AI hardware market, at potentially more favourable pricing. This rivalry extends beyond just the foundry business into memory and packaging, where both companies are significant players.

    Implications for Asia’s Electronics Supply Chain

    For Asia’s electronics and consumer tech sectors, this dynamic is crucial. Many consumer devices, from smartphones to smart home hubs, increasingly rely on AI capabilities, which in turn depend on advanced chips and memory. A shift in pricing power or supply availability from major manufacturers like TSMC and Samsung can ripple through the entire supply chain, affecting component costs and product development timelines for brands across the region. RetailNews Asia tracks how such foundational shifts in manufacturing impact the availability and pricing of critical components for major electronics brands and, ultimately, the consumer market.

  • Samsung Foundry drops further behind TSMC after disasterous Q4

    Samsung Foundry drops further behind TSMC after disasterous Q4

    TSMC is the largest contract chip foundry in the world. Looking at its roll call of customers, you can see why it is number one. After all, the Taiwan-based foundry manufactures chips based on designs submitted by companies such as Apple, Nvidia, AMD, MediaTek, Qualcomm, and Broadcom. Data compiled by TrendForce that The Chosun Daily published says that during the fourth quarter of 2024, TSMC’s market share rose 2.4 percentage points to 67.1%.
    That means approximately two out of every three chips produced during the last three months of the year were built by TSMC. As for second-place Samsung Foundry, its market share dropped 1 full percentage point from 9.1% to 8.1% during the same quarter. Sammy’s foundry has had to deal with poor yields on its advanced chips, which this year resulted in the use of Qualcomm’s Snapdragon 8 Elite for Galaxy application processor (AP) on every Galaxy S25 series unit.
    A repeat could be in store for next year’s Galaxy S26 line unless Samsung Foundry can hike its yield on 2nm production from the current 30% before mass production of the Exynos 2600 AP begins. If there is enough of an improvement, the Exynos 2600 SoC will power the Galaxy S26 and Galaxy S26+ in all markets except for the U.S., China, and Canada.
    But we digress. The gap in market share between TSMC and Samsung Foundry increased sequentially from 55.6% in Q3 to 59% in Q4. TrendForce says that TSMC is benefiting from the AI buzz as demand for Nvidia’s GPU chips used for AI training, inference (predictions and decisions), and computing picks up. Those chips are made by TSMC.
    Also helping TSMC is the demand for high-end smartphone chips including those used to power the iPhone, and Samsung’s flagship Galaxy S and foldable Galaxy Z models. Starting this year the new Tensor G5 AP, designed by Google from the ground up, will be built by TSMC using its third-generation 3nm node (N3P) and deployed in the Pixel 10 line. That is the same node that TSMC will use to build Qualcomm’s Snapdragon 8 Elite 2 for next year’s Galaxy S26 Ultra.
    You might be wondering which foundry is in third place behind Samsung Foundry. That would be China’s largest foundry, SMIC. While the latter appears to be handling the production of Huawei’s APs, it also could be making some money building Huawei’s Ascend AI accelerators. Long-time PhoneArena readers probably are aware that thanks to U.S. sanctions, SMIC cannot purchase extreme ultraviolet lithography machines preventing SMIC from building chips using a process node more advanced than 7nm.
    TSMC generated $26.85 billion in revenue during Q4 2024, a gain of 14.1% year-over-year. Samsung Foundry’s Q4 revenue declined 1.4% to $3.26 billion.
    Lees ook:

  • Apple chip builder TSMC not yet ready to move to next-gen lithography system

    Apple chip builder TSMC not yet ready to move to next-gen lithography system

    Back in December, Dutch tech giant ASML shipped its first High-NA Extreme Ultraviolet (EUV) lithography machine (EXE:5000) to Intel. The $400 million machine will take us to the next tier of chip production with a process node of 2nm and less. The original EUV machines, also made by ASML, were necessary to get chip foundries to manufacture components under 10nm. A lower process node means smaller transistors, meaning more can fit inside a chip. The higher a chip’s transistor count, the more powerful and/or energy efficient it is.

    The reason that the EUV machine is so important is that it prints circuitry patterns on silicon wafers that are thinner than human hair. This is required when you are building a chip with billions of transistors inside. The 7nm A13 Bionic SoC, used to power 2019’s iPhone 11 series, contained 8.5 billion transistors. The 3nm A17 Pro used to run the iPhone 15 Pro and iPhone 15 Pro Max sports 19 billion transistors.

    The 8 nm resolution of the EXE:5000 means that chipmakers can pack more transistors into a single chip. Smaller transistors are more energy efficient – that means the chips will be able to do more with less.-ASML

    The new High-NA EUV machine has a 0.55 numerical aperture (NA) lens giving it a resolution of 8nm compared to the current machines’ resolution of 13nm (.33 NA). This means that the new machines can print transistors 1.7 times smaller resulting in transistor densities 2.9 times larger with a single exposure. The result? More powerful or energy-efficient chips. The new machines can also print 185 wafers per hour, rising to 220 by 2025. That compares to the 160 wafers per hour that can be printed with the current EUV machines.

    The current Low-NA EUV machines can produce the same resolution but only after two exposures are made using double patterning. However, there are risks with double patterning including longer production times, and increased risk that a defect will occur. It can also lead to performance variability among manufactured chips.

    ASML is quick to tell you what can go wrong with double patterning since it would prefer foundries spend the bigger bucks on the newer machines. But TSMC’s N3B process node, which supposedly relies on double patterning, was used to manufacture the A17 Pro application processor used on all of those iPhone 15 Pro and iPhone 15 Pro Max units made by Apple, and the M3 chip used to power high-end Macs. ASML says that its clients are just now doing their research into the High-NA EUV.

    Financial institution China Renaissance (via

    Eventually, TSMC, Samsung Foundry, and others will have to join Intel and start making the investments needed to move to the High-NA EUV. That time might end up being soon with TSMC and Samsung looking to start 2nm production in 2025 moving to 1.4nm production in 2027.

    Lees ook:

  • TSMC shows off 2nm chip prototype to Apple

    TSMC shows off 2nm chip prototype to Apple

    TSMC is the largest foundry in the world and Apple is its biggest customer. Apple was able to reserve most of TSMC’s initial 3nm production for the A17 Pro chipset which is being used to power the iPhone 15 Pro and iPhone 15 Pro Max. To keep things simple, just remember that as the process node number gets lower, the transistors used by a chip get smaller in size which means that more of them can fit into the small space inside an SoC. The higher a chip’s transistor count, the more powerful and/or energy-efficient it is.

    Looking at the iPhone as an example, the iPhone 11 series was powered by the 7nm A13 Bionic which carried 8.5 billion transistors. The 3nm A17 Pro has 19 billion transistors. And since it takes a few years to build the fabs and order the machinery needed, we’ve known for some time that TSMC is working toward 2nm chip production. Apple has already had a chance to check out 2nm prototypes built on TSMC’s N2 node. The report noted that another huge TSMC customer, NVIDIA, also was able to view the 2nm prototype chip.

    TSMC previously said that it would start 2nm volume production in 2025 and in a statement to The Financial Times, the company said that it is “progressing well and on track for volume production in 2025, and will be the most advanced semiconductor technology in the industry in both density and energy efficiency when it is introduced.” Assuming that there are no delays, the iPhone 17 Pro and iPhone 17 Pro Max could be the first Apple handsets to be powered by a 2nm application processor (AP), possibly the A19 Pro.

    At 2nm, TSMC will debut its new Gate-all-around (GAA) transistors which cover the channel on all four sides reducing current leaks and delivering gains in power efficiency. Samsung Foundry already uses GAA with its 3nm production.

    Speaking of Samsung Foundry, the latest word is that Qualcomm will be switching from TSMC to Samsung to produce the 3nm Snapdragon 8 Gen 5 AP in 2025. Qualcomm left Samsung Foundry in 2022 after the latter reported low yields building the Snapdragon 8 Gen 1. TSMC took over to produce the Snapdragon 8+ Gen 1 and has had the business ever since, although that seems likely to change in 2025. Samsung Foundry’s yield is about 60% for the most basic of 3nm production. The yield is expected to drop when building APs for smartphone.

    At a 60% yield, 40 out of every 100 dies cut from a wafer fails to meet quality control. And since the foundry’s customers are usually responsible for the cost, yield is a major factor when determining which foundry a chip designer does business with.

    Samsung says that it is ready for 2nm production. “We are well-equipped to set up for SF2 mass production by 2025,” Samsung said. “Since we were the first to take the leap and transition to GAA architecture, we are hoping the progress from SF3 to SF2 will be relatively seamless.” While Apple and NVIDIA were able to get a look at their future, TSMC is putting together the final list of its 3nm and 2nm clients according to DigiTimes.

    Intel is also looking to grab some contract business away from TSMC and Samsung Foundry. Intel’s next-gen 18A node (1.8nm) could take process leadership away from the pair and it is offering chip design firms free test production. Intel could shake up the market. Longtime TSMC customer AMD said in July that it would “consider other manufacturing capabilities.”

    Leslie Wu, chief executive of consulting firm RHCC, said that companies looking to manufacture their chip designs using a 2nm node might start having their chip production spread across multiple foundries. With the concern about China invading Taiwan always something to worry about, Wu said, “It’s too risky to rely on TSMC solely.” Yet Apple has been relying solely on TSMC for years.

    Lees ook: