The global semiconductor landscape is undergoing a fundamental shift, moving from a focus on raw training power to the practical complexities of large-scale deployment. In an in-depth interview, Wallace Kou, President and CEO of Silicon Motion, detailed how the generative AI has evolved beyond its initial stages. While the market's early gaze was fixed almost exclusively on NVIDIA's GPUs, the High Bandwidth Memory (HBM), and the CoWoS advanced packaging technology, Kou argues that the industry is now entering the "Inference" era that is turning previous under-estimation about storage's importance on their head.The Shift from Training to InferenceThe turning point for this realization occurred during the NVIDIA GTC conference in March 2026. CEO Jensen Huang unveiled the Vera Rubin architecture, a move that signaled a massive spike in demand for NAND flash memory. During the initial AI boom, the industry was preoccupied with training massive models, a process that relies heavily on the lightning-fast throughput of HBM. However, as these models move into the inference phase - where they are actually used by end-users to generate content or solve problems - the access to context, historical data, and massive datasets storage become the primary bottleneck.Kou notes a dramatic shift in market sentiment. Only two years ago, storage was often an afterthought in the AI conversation; today, it is a critical scarcity. "There is currently not a single global cloud service provider or major smartphone manufacturer whose demand for DRAM and NAND is being fully satisfied," Kou observed. This supply-demand gap has triggered a financial windfall for storage module manufacturers and memory giants, with some stock prices skyrocketing up to tenfold as the market reacts to persistent shortages and rising prices.Technical Paradigm Shift: CMX and the Infrastructure of ThoughtAt the heart of this transition is a new architecture introduced by NVIDIA: the CMX Context Memory Storage platform. This architecture is designed specifically to handle the "KV Cache" (Key-Value Cache), which allows AI models to remember the context of a conversation or a complex task during the inference process.The hardware requirements for the CMX architecture are staggering in their scale and technical demands. Each individual Rubin GPU requires 16TB of dedicated storage to function effectively within this framework. At a system-level scale, a single NV72 Vera-Rubin setup can demand more than 1 Petabyte, or 1,000 Terabytes, of total storage capacity. Beyond mere capacity, the CMX architecture facilitates direct GPU access to storage, a feature that bypasses traditional latency bottlenecks and ensures that AI inference remains fluid and responsive.While this creates a massive commercial opportunity for the storage industry, it also places an unprecedented strain on NAND production. Kou emphasizes that this is not just a cloud-based phenomenon. The explosion of Edge AI - AI processed locally on devices - is further complicating the supply chain. For instance, driven by major players like Meta, the market for smart glasses is expected to reach 60 million units this year. These wearable devices require high-performance embedded storage, creating a secondary front in the war for NAND capacity.Silicon Motion's Role: Solving the QoS BottleneckAs the world's leading NAND controller maker, Silicon Motion sits at the intersection of these competing demands. The primary technical challenge in modern AI environments is maintaining Quality of Service (QoS). In a multi-tenant cloud environment, where multiple GPUs are accessing shared storage simultaneously for different inference tasks, data transfer speeds can often fluctuate or drop.To solve this, Silicon Motion has deployed its proprietary PerformaShape technology. This technology ensures that even under heavy, concurrent workloads, the transmission speed remains stable. By stabilizing these data flows, Silicon Motion has positioned itself as an "indispensable stabilizer" in the AI ecosystem.Beyond data path optimization, Silicon Motion is also extending its role into system-level infrastructure by providing enterprise-grade boot drives for leading AI GPU, TPU, and DPU platforms, ensuring system reliability and fast initialization at scale.The Crisis of Imbalance: Kou's "Capacity Persuasion" EffortsDespite the record-breaking revenues, Kou is deeply concerned about the "shadows" lurking behind this prosperity. The current memory market is suffering from a dangerous imbalance. To maximize profits and satisfy the insatiable hunger of AI cloud giants, major manufacturers like Samsung, SK Hynix, and Micron are funneling the majority of their capital expenditure (CAPEX) into HBM and DDR5 production.This strategic pivot has effectively "squeezed" the production capacity available for standard NAND flash. Kou warns that this "AI squeezing effect" could lead to a collapse in traditional sectors. Over the past eight months, Kou has embarked on a global mission, meeting with leaders at Samsung, SK Hynix, Kioxia, SanDisk, YMTC, and Micron. His message is one of "capacity persuasion": he is urging these giants to reserve a portion of their production lines for the automotive, PC, and smartphone industries."If these foundational industries break because they cannot find parts, Edge AI will have no 'soil' to grow in," Kou warned. He believes that a total focus on the high-margin AI server market could eventually backfire, destroying the broader technology ecosystem that supports AI development.A Stabilizing Strategy: From Cloud to EdgeSilicon Motion is positioning itself as the "transition enabler" for an industry in flux amid an expected 2–3 year supply shortage. As NAND manufacturers concentrate their internal resources on AI-driven initiatives, they are increasingly outsourcing non-core and mainstream projects, such as PCIe Gen5 controllers and embedded solutions. In this shift, Silicon Motion has emerged as a preferred partner to fill the resulting gap.At the same time, as rising prices weigh on demand in the PC and smartphone markets, the company is helping customers pivot toward automotive and AIoT applications, including rapidly growing segments such as smart glasses, which are seeing a surge in shipments this year.One of the most critical areas is the automotive sector, where Silicon Motion has spent a decade building a presence. While memory giants might see automotive requirements as "niche" or low volume compared to AI servers, Kou views them as essential to global stability. When major OEMs consider abandoning these specialized demands due to capacity constraints, Silicon Motion steps in to ensure the global automotive supply chain does not grind to a halt."We are not just looking for a surge in revenue; we want to fulfill our responsibility to the industry," Kou said. By providing stable controllers and storage solutions for AIoT and automotive applications, Silicon Motion is effectively repairing the cracks in a fractured global supply chain.Future Outlook: 2027 and BeyondThe current supply-demand imbalance is not a temporary glitch but a structural reality that Kou expects to persist until at least late 2027 or 2028. Several factors make it nearly impossible to add capacity quickly, for example, land acquisition is increasingly difficult. The lead time for building specialized cleanrooms and procuring critical equipment now exceeds one year.Kou predicts that while the DRAM shortage might begin to ease by the end of 2027, the relief for NAND will likely come even later. In this high-pressure environment, Silicon Motion's role as a key stabilizing force becomes increasingly important.Particularly in emerging sectors such as smart IoT and automotive applications, Silicon Motion delivers reliable controller and storage solutions, filling the vacuum left by production shifts at major manufacturers or by projects lacking sufficient engineering support.By helping global clients navigate the complexities of geopolitics and capacity wars, Silicon Motion aims to ensure that the AI revolution leads to a steady, sustainable future rather than a chaotic collapse of the broader tech industry.AI inference boom fuels supply-demand imbalance until 2027-2028, says Wallace Kou. Credit: Silicon Motion

With demand for AI edge computing continuing to grow, Retronix announced that its Sparrow Hawk AI edge computing platform has officially entered mass production and is now available to the global market. The platform is powered by the automotive-grade SoC R-Car V4H from Renesas Electronics, delivering up to 30 TOPS (Dense) AI inference performance. Sparrow Hawk is designed to support a wide range of edge AI applications, including robotics, smart manufacturing, and other intelligent edge systems.According to Retronix, Sparrow Hawk integrates high-performance AI computing capabilities with a comprehensive I/O expansion architecture, enabling diverse edge computing scenarios such as computer vision, industrial automation control, and smart device development. The platform provides a wide range of interfaces including PCIe, USB 3.0, DisplayPort, CAN-FD, AVB Ethernet, and MIPI CSI/DSI. It also features an expansion connector compatible with industry-standard accessories, allowing developers to flexibly integrate peripheral modules and accelerate AI application development and deployment."We are pleased to see the global launch of Retronix's Sparrow Hawk Edge AI Computing Platform, powered by the automotive-grade R-Car V4H SoC," said Aish Dubey, Vice President and Head of the SoC Division at Renesas Electronics. "We value our collaboration with Retronix and look forward to continuing our work together to bring more innovation to the edge AI ecosystem."Regarding product compliance, Sparrow Hawk has obtained several international certifications, including FCC, CE, UKCA, and VCCI. These certifications indicate that the product meets international standards for electromagnetic compatibility (EMC), safety regulations, and other regulatory requirements, enabling Sparrow Hawk to be legally marketed and deployed in major regions including the United States, the European Union, the United Kingdom, and Japan.Sparrow Hawk was developed with an Upstream First software strategy that helps reduce entry barriers for developers. To meet diverse application needs, it fully supports two major software distributions. The Debian BSP is primarily aimed at industrial applications, providing a familiar software installation experience, while the Yocto BSP continues the high performance and reliability DNA of the R-Car series in the automotive field. Developers can access source code, Startup Guides, and detailed technical tutorials directly through the R-Car Open Source Community GitHub page. The community has established FAQ and Q&A forums, allowing global developers to communicate with experts in real-time and accelerate the transition from prototype to production.For hardware configuration, Sparrow Hawk is available in 8GB and 16GB LPDDR5 memory versions. The Basic Kit includes the Sparrow Hawk SBC, while the Complete Kit adds a 32GB microSD card, a fan HAT and a debug serial cable for a ready-to-use development setup.The Sparrow Hawk single-board computer and development kits are now officially available through Shimafuji Electric, with broader availability planned through major electronic component distributors such as Mouser Electronics and Digi-Key Electronics. Retronix Technology will work closely with Renesas Electronics to provide comprehensive technical support for developers and customers worldwide.For more technical details, please visit: Retronix Official Website.

Global Unichip Corp. (GUC), the Advanced ASIC Leader, will showcase Jotunn8, a next-generation data center AI inference processor developed by VSORA, at the TSMC Europe Technology Symposium.This exhibition highlights the successful collaboration between GUC and VSORA, combining advanced AI architecture with leading-edge ASIC implementation and packaging technologies to deliver a high-performance, scalable inference solution for data center workloads.Enabling advanced AI inference through close collaborationJotunn8 is VSORA's flagship AI inference processor, purpose-built for inference workloads. It delivers ultra-low latency and very high throughput by addressing the memory wall bottleneck, enabling cost-efficient deployment of large-scale AI models.GUC provided comprehensive turnkey ASIC services for Jotunn8, managing the full implementation from netlist to manufacturing. The project demonstrates GUC's expertise in complex system integration, including: Advanced chiplet architecture design and integration, High-bandwidth memory integration with HBM3E PHY and controller, 2.5D die-to-die connectivity using GUC's 17.2 Gbps GLink-2.5D interconnect, Advanced packaging implementation leveraging TSMC CoWoS-S technology (3x reticle size), Implementation on TSMC's 5nm process node,Full-system co- optimization of signal, power, and thermal integrity (SI/PI/TI), Power and IR optimization to enhance overall system efficiency.This collaboration underscores GUC's capability to deliver complex AI and HPC ASICs with optimized performance, power, and scalability.Leveraging a strong ecosystem with TSMCJotunn8 also reflects the strength of GUC's long-standing collaboration with TSMC. As a key partner in TSMC Open Innovation Platform (OIP) ecosystem, GUC enables customers to efficiently adopt advanced process nodes and packaging technologies, bridging innovative architectures with leading-edge manufacturing.By leveraging TSMC's advanced CoWoS packaging and 5nm process technologies,GUC ensures high-quality silicon execution and a reliable path to production for demanding AI applications."We are proud to showcase Jotunn8 at the TSMC Europe Technology Symposium. This achievement highlights our strong collaboration with VSORA and demonstrates GUC's ability to bring complex AI processors to silicon using advanced process and packaging technologies," — said Patrick Wang, Senior Vice President and Chief Revenue Officer at GUC."Jotunn8 is designed to unlock a new level of efficiency and scalability for AI inference. Our collaboration with GUC has been key in successfully translating this architecture into a highperformance silicon solution,"— said Khaled Maalej CEO of VSORA.Driving next-generation data center AIThrough this collaboration, GUC and VSORA address the growing demand for high-performance,energy-efficient AI inference in hyperscale data centers. Jotunn8 enables new levels of throughput and latency optimization, supporting large-scale AI deployment across cloud and enterprise environments.

Global Unichip Corp. (GUC), the Advanced ASIC Leader, today announced a strategic technical collaboration with Wiwynn, an innovative cloud IT infrastructure provider for data centers. This collaboration integrates GUC's flagship SoC design and 2.5D/3D advanced packaging with Wiwynn's expertise in rack-scale system integration, liquid cooling and optical interconnect. Together, the collaboration enables hyperscale customers to transition more efficiently from silicon definition to deployment-ready AI infrastructure.AI clusters continue to scale in performance, bandwidth and power density, hyperscalers must increasingly evaluate silicon, packaging, interconnect, thermal and rack-level design choices much earlier in the development cycle. Through this collaboration, GUC and Wiwynn are aligning key technology pillars, including leading-edge ASIC implementation, 2.5D/3D advanced packaging, optical I/O, power delivery, thermal architecture, manufacturability, serviceability and rack-scale integration. By addressing these factors holistically at the outset, the collaboration partners aim to reduce integration complexity, improve development efficiency and accelerate the transition from silicon-ready innovation to system-ready AI infrastructure."As AI infrastructure evolves beyond chip-level optimization and scale-up networks push the limits of conventional electrical interconnects, close alignment across silicon to system architecture become critical," said Aditya Raina, Chief of Marketing of GUC. "By collaborating with Wiwynn, we are helping hyperscale customers evaluate critical system-level tradeoffs earlier, integrating optical I/O to deliver the bandwidth and power efficiency required for next generation AI systems. This partnership establishes a more practical, holistic path from flagship ASIC development to deployable, rack-scale AI infrastructure."With deep expertise across board-level innovation, rack-scale integration, and manufacturing, Wiwynn effectively bridges semiconductor innovation with data center deployment," said Tony Wen, Vice President at Wiwynn. "Together with GUC, we are enabling a comprehensive silicon-tosystem approach that delivers scalable, efficient and serviceable AI infrastructure tailored for nextgeneration hyperscale environments."For more information, please visit.

Taiwan Mobile today held a Mac launch event to announce that Mac products are now officially available. Centered around the theme "Small Bite. Big Possible.", Taiwan Mobile has integrated MacBook Neo, M5-powered MacBook Air and MacBook Pro, as well as Mac mini, becoming the first telecom operator in Taiwan to deliver a full-scale Apple ecosystem.Beyond the simultaneous launch across myfone retail stores, myfone online shop, customer service channels, enterprise divisions, and momo.com, Taiwan Mobile has introduced dedicated "Mac Experience Gateways" across 100 select myfone retail stores. These areas are supported by over 1,000 professionally trained AI Tech Consultants who provide hands-on demonstrations and personalized guidance nationwide.At the launch event, a live demonstration showcased Taiwan Mobile’s evolution from device sales to integrated experiences. The showcase highlighted both the cross-device Apple ecosystem experience and Taiwan Mobile's own advantages-5G hotspot sharing, home Wi-Fi broadband, diverse purchasing plans, and OP Tech Life services-illustrating how Taiwan Mobile delivers one-stop solutions across network, device, and content services, tailored to different user needs, empowering customers to maximize their digital productivity in the AI era.Tony Lin, Chief Consumer Business Officer at Taiwan Mobile, stated that as the "AI Inquiry Era" emerges, the challenge for consumers is no longer about choosing devices, but about how to effectively harness and integrate technology into everyday life. Through today's launch event demonstrations, Taiwan Mobile showed that retail stores are no longer just points of sale, but have evolved into integrated service hubs. myfone retail stores continue to strengthen their consultative service approach, upgrading into "Experience Gateways," where AI Tech Consultants guide users in understanding how devices connect and work together-bringing "Big Possible" to life.Tony Lin, Chief Business Officer at Taiwan Mobile, at the Mac Launch Event. Credit: Taiwan MobileLin further noted that Taiwan Mobile integrates 14 value-added services, including Apple One, streaming content, and cloud gaming. With select plans, customers will also receive a one-year free membership to the "GeForce NOW powered by Taiwan Mobile" Ultra Plan Hourly Subscription, enabling them to enjoy AAA titles on Mac and further expanding its entertainment and productivity value.Live from the Launch Event: 3 Real-Life Scenarios Across Students, Professionals, and CreatorsStudents: MacBook Neo at $0 - A Seamless Learning ExperienceThe first scenario demonstrates a student’s learning journey, from attending classes to reviewing course materials, addressing the challenge of fragmented information and inefficient organization. Through Taiwan Mobile's device plans, users can easily get started with the latest MacBook Neo and step into the Apple ecosystem. By using iPhone Air to capture whiteboard content and extract or translate key information, and iPad Air to take notes during class, students can seamlessly sync everything to MacBook Neo for further review, creating a smooth and connected workflow for capturing, organizing, and learning.After completing schoolwork, students can transition into leisure and entertainment. With GeForce NOW powered by Taiwan Mobile, there is no need to install large game files, while game progress can be saved across devices through cloud saves, enabling access to AAA titles via cloud gaming. From productive learning during the day to relaxing entertainment at night, MacBook Neo paired with Taiwan Mobile's low-latency, ultra-fast 5G and Wi-Fi delivers everything students need for a modern digital lifestyle.Professionals: Unlimited 5G Hotspot Sharing Powers Seamless Work on MacBook AirThe second scenario focuses on professionals, highlighting how to maintain efficiency amid fragmented information and the growing demand for mobile productivity. It also showcases Taiwan Mobile's NT$1,599 monthly plan with unlimited 5G hotspot sharing.With MacBook Air paired with iPhone and Apple Watch, users can stay productive anytime, anywhere. Whether reviewing reports during the commute, delivering presentations on the high-speed rail, or joining last-minute online meetings, they can rely on stable connectivity and smooth performance-turning every location into a mobile workspace.Creators: OP Tech Life and Double Play Plan Enable a Seamless Creative Workflow with Mac miniThe final scenario focuses on creators, demonstrating how OP Tech Life and Double Play Plan, which integrates mobile connectivity and home Wi-Fi, address the challenges of fragmented processes across shooting, organizing, and post-production.When shooting on the go, iPhone 17 Pro Max captures 4K video, while 5G connectivity enables real-time content transfer and processing. Back in the studio, high-speed broadband supports a smooth continuation of the creative process. With AirDrop, files can be quickly transferred and seamlessly integrated with Mac mini and other devices, allowing the entire workflow, from shooting to editing and post-production, to be completed efficiently. This empowers creators to maintain creative momentum and maximize productivity without interruption.Four New "OP Tech Life" Bundles: Elevating Mac Across All ScenariosTo further enhance the Mac experience, Taiwan Mobile's "OP Life" series introduces four curated bundles, providing users with a complete, out-of-the-box solution for various use cases.AI All-in-One Smart Laptop BundlesDesigned for the MacBook Neo, MacBook Air, and MacBook Pro, these bundles pair Mac notebooks with premium Logitech peripherals, including the Lift for Mac Ergonomic Vertical Mouse or the MX Master 4 for Mac Wireless Mouse. These combinations cater to entry-level users, advanced creators, and high-productivity professionals alike.MacBook Neo Bundle: Available for $0 with a NT$1,399 (48-month) plan.MacBook Air & MacBook Pro Bundles: Starting at $15,900 and $34,900 respectively, with NT$1,399 and NT$2,699 (48-month) plans.AI All-in-One Mini Desktop BundleFor desktop users, the Mac mini is paired with the Logitech K580 Wireless Keyboard to meet the demands of home offices and multitasking. This versatile setup is available for $0 with a NT$1,399 (48-month) plan.MacBook Neo AI All-in-One Bundle: $0 with NT$1,399 plan (48 months). Credit: Taiwan MobileOver 1,000 AI Tech Consultants Lower Barriers to Technology Use Across Four Key AreasIn addition to its multi-channel sales rollout, Taiwan Mobile has established dedicated Mac demonstration zones across 100 select myfone retail stores. Nationwide, more than 1,000 AI Tech Consultants are available to support customers. Covering four key areas across product expertise, device connectivity, customer service, and problem-solving, they help lower the barriers to cross-device usage.By integrating 5G and Wi-Fi broadband connectivity, device selection, and value-added content services, Taiwan Mobile aims to enable customers to activate and maximize the value of the Apple ecosystem from the moment they step into the store.Apply for Select Plans for a Chance to Win a Gold Coin; Online Shop Offers Additional Chances to Win MacBook Neo and iPhone 17eFrom now until May 31, 2026, customers who subscribe to, switch to, or renew eligible plans priced at NT$1,399 or above, bundle with select Mac products, and complete registration on the official website by June 2 will receive one entry for a chance to win a gold coin. Customers who subscribe to eligible plans priced at NT$1,599 or above, renew or upgrade their plans, add annual subscriptions for streaming services, or subscribe to "Double Play Plan" with 300 Mbps or higher broadband can receive up to five entries.In addition, customers who apply for designated plans or purchase Mac products via the myfone online shop will have the chance to win MacBook Neo, with an extra draw for iPhone 17e available until April 30. Renewal customers can receive up to NT$1,000 in myfone e-voucher rewards, with an additional NT$200 for social sharing. New subscriptions or number portability applications are also eligible for NT$500 in myfone e-vouchers.Bundle Mac with NT$1,399+ monthly plans by May 31 to win gold coins. Credit: Taiwan MobileSubscribe to Eligible 5G Plans to Enjoy a One-Year Free "GeForce NOW Powered by Taiwan Mobile" Ultra Plan Hourly SubscriptionGet MacBook Neo at $0 with Taiwan Mobile's NT$1,399 monthly plan. Credit: Taiwan MobileCustomers who subscribe to Mac products with designated plans can receive trade-in discounts of up to NT$32,100, along with additional savings of up to NT$5,000 for VIP renewals. MacBook Neo is available at $0 under eligible plans. Furthermore, using the "Open Possible" Co-branded Card for purchases at myfone stores earns 2% cashback, alongside up to 3.5% cashback on telecom bills. Customers will also enjoy a range of value-added services, including a two-month free trial of Apple One, NT$50 App Store billing credits, and up to four months of YouTube Premium. In addition, a 12-month free membership to "GeForce NOW powered by Taiwan Mobile" Ultra Plan Hourly Subscription and a six-month free trial of the OP Scam Buster advanced plan.Taiwan Mobile further offers exclusive benefits such as a three-month free trial of MyVideo, Business Today app, CommonWealth Magazine Digital Subscription, and WeMo PASS, along with NT$50 ride coupons for daily rental services, providing an integrated experience across entertainment, reading, and smart lifestyle services.For streaming services, Taiwan Mobile offers a variety of discounted plans, including Prime Video at NT$89/month (originally NT$169), HBO Max Standard Plan (annual subscription, monthly payment) at NT$160/month (originally NT$220), and Disney+ 24-month add-on plan at NT$199/month (originally NT$285), as well as the Taiwan Mobile exclusive Multi-Streaming Package at up to 44% off. Additionally, customers on the "Double Play Plan" 1Gbps designated plan can enjoy selected home appliances worth up to NT$10,000 at no cost, or up to 10 months free on the Multi-Streaming Package, plus a complimentary 3-month KKBOX trial-delivering a more complete digital entertainment experience.NT$100 Off for Every NT$1,000 Spent on Selected Accessories for MacTaiwan Mobile is also introducing promotions on Mac-related accessories, covering both Apple original products and selected third-party options to enhance the overall user experience. Apple accessories include Magic Mouse with USB-C, Magic Keyboard with USB-C, USB-C Digital AV Multiport Adapter, 240W USB-C Charge Cable, 40W Power Adapter, and 70W USB-C Power Adapter.Selected third-party accessories include laptop sleeves, the Momax ONELINK 8-in-1 USB-C multi-function adapter, magnetic privacy screen protectors designed for MacBook 13.6-inch and MacBook Neo, as well as magnetic aluminum stands, providing a wide range of options for different user needs.

Global Unichip Corp. (GUC), the advanced ASIC leader, today announced the successful demonstration of a 12 Gbps HBM4 IP platform implemented on TSMC's 3nm process technology at the Partner Pavilion of the TSMC 2026 North America Technology Symposium. The platform features GUC's in-house full functional HBM4 Controller and PHY IP, integrated with a partner's HBM4 memory, and used TSMC's industry-leading CoWoS advanced packaging technology.GUC's previous-generation HBM3E PHY and Controller, deployed in customers'3nm products, have achieved speeds 15% above specification in production. JEDEC continues to drive an aggressive HBM roadmap, increasing memory throughput and capacity while further doubling the data bus width in HBM4. Comparing with HBM3E, GUC's HBM4 IP delivers 2.5x bandwidth, while improving power efficiency by 1.5x and area efficiency by 2x.In line with GUC's previous HBM, GLink, and UCIe IP solutions, the HBM4 IP integrates proteanTecs' interconnect monitoring solution to provide high visibility for PHY testing and characterizing, while enhancing in-field performance and reliability for end products.To address the growing demand for 3DIC architectures, GUC's HBM4 PHY also supports a face-upconfiguration, enabling integration with TSMC's SoIC face-to-face technology. The PHY macrointegrates TSVs for PHY's I/O signals, power, and ground connections, and also reserves TSVs forpower feedthrough to the top die, supporting the power distribution requirements of the upper logic die."We are proud to be the first company to demonstrate a 12 Gbps HBM4 IP to customers at theTSMC Symposium" said Igor Elkanovich, CTO of GUC. "Together with GUC's UCIe and GLink-3D IPs, we offer a complete 2.5D/3D IP solution for modern 3.5D system architectures, includingTSMC SoIC-X on CoWoS."For more information,please visit our website.12 Gbps Eye Diagram. Credit:Guc

AI EXPO Taiwan 2026, the premier annual technology event in Asia, concluded with record-breaking success at the Taipei Expo Dome. Organized by DIGITIMES in collaboration with IC Radio Broadcasting, College of Intelligent Computing and Quantum Information of Chung Yuan Christian University, the National Development Council, and the Taipei Expo Foundation, the event’s fifth anniversary marked a historic milestone under the theme "AI·X: Infinite Cross-Domain."AI EXPO Taiwan 2026 saw a staggering 93% year-on-year increase in registrations, surpassing 50,000 attendees. Featuring 300 premier brands and more than 350 enterprise AI solutions and autonomous agents, the event solidified Taiwan's position as the mission control of the global AI supply chain.National Strategy Meets Urban GovernanceThe opening day was graced by Vice President Bi-khim Hsiao and Taipei Mayor Wan-an Chiang, signaling that AI development has ascended to the level of national security and strategic priority.Colley Hwang, Chairman of DIGITIMES and IC Broadcasting, highlighted that Taiwan's electronics industry has reached a staggering output of $1.18 trillion USD. With over 90% of the world's AI servers now manufactured by Taiwanese firms, the nation has solidified its indispensable role at the core of the global supply chain, marking what is currently the strongest era in Taiwan's industrial history.Mayor Chiang highlighted Taipei's leadership as the first city in Taiwan to implement AI governance guidelines, noting that AI-powered municipal services have boosted the efficiency of the 1999 Citizen Hotline by 90%, providing a blueprint for AI-driven urban management.The event also demonstrated powerful momentum through the active participation of various local governments. Chiu Chao-mei, Director of the New Taipei City Youth Bureau, led a cohort of promising startups to showcase their innovations. Meanwhile, Chang Cheng, Director of the Taoyuan City Economic Development Bureau, engaged in high-level exchanges regarding industrial upgrading. Additionally, the Hsinchu County Government organized a professional delegation for an on-site study tour. Together, these initiatives highlight the seamless integration between Taiwan’s strategic tech corridor and the burgeoning AI ecosystem.Market Dynamics: Inference, Autonomous Agents, and Physical AIThe DIGITIMES Research Center revealed a critical shift in computing power toward inference. Global high-end AI server shipments (equipped with HBM) are projected to reach 1.914 million units, representing 67.5% year-on-year growth, as supply chain bottlenecks for components and system assembly continue to ease.Industry leaders identified 2026 as the "Year One" for AI killer applications:AI Agents: Microsoft Taiwan and IBM Taiwan showcased the transition to an era where humans lead and agents execute. Solutions ranging from Dell's "AI Factory" to 91APP's retail agent platform signal that AI has moved beyond proof of concept (PoC) into full-scale commercialization.Physical AI: Advantech and Innodisk demonstrated the integration of Edge AI and robotics, enabling real-time computing and functional safety in industrial environments.Quantum Frontiers:Prof. Ching-Ray Chang, Director of CYCU Quantum Information Center, joined experts from Google and NVIDIA to discuss "Quantum-HPC Integration" and warned of potential "Q-Day" security risks, urging organizations to accelerate adoption of post-quantum cryptography (PQC).Credit: DIGITIMESCross-Domain Synergy: From K-POP to Humanistic ValuesThe expo transcended technical boundaries by merging the entertainment economy with philosophical reflection. Lim Seul-ong, a member of the legendary Kpop group 2AM, appeared as Chief Creative Officer of INIP to discuss how AI and XR are reshaping K-POP's business models.In the realm of content, the "Creator Influence Awards" - jointly organized by DIGITIMES, IC Broadcasting, and DailyView - honored PanSci as "Creator of the Year," recognizing the vital role of scientific literacy in the AI era.The event concluded with a keynote by renowned artist Shih-Yi Yang titled: "This is Not AI: Seeing the Beauty of the Real World." Yang offered a poignant reminder: "Salt crystallizes through the sun; humans find clarity through stillness." He urged the audience to reconsider the value of time in an age where AI maximizes technical efficiency.Looking Forward: Leading the Global WaveFrom "Standing in the Future" in its first year to "Infinite Cross-Domain" in its fifth, AI EXPO Taiwan has evolved into a global catalyst for innovation. AI EXPO Taiwan 2026 proves that Taiwan remains at the heart of AI innovation and global influence. We thank our international partners for their incredible support. We look forward to welcoming you back in 2027 as we continue to push the boundaries of what's possible.

The "Mini SSD Strategic Seminar" convened recently to strengthen collaboration and accelerate ecosystem development following the successful "Mini SSD Ecosystem Application Seminar" held in Shenzhen on January 26.Leading companies across the industry chain, including Intel, BIWIN, Longcheer, BYD Electronics, Luxshare, Emdoor, Weibu, JWIPC, CYX Industrial, MINISFORUM, Decenta, and SIXUNITED, reached a strategic consensus to jointly build the Mini SSD industry ecosystem, marking a significant step toward large-scale adoption of this emerging storage form factor.As the first batch of core builders of the Mini SSD ecosystem, the initial partners will gain early access to patent licensing benefits, as well as the synergies created through ecosystem collaboration. This milestone signals that Mini SSD is evolving from a standalone product innovation into a collaborative, win-win industry ecosystem.Global Recognition: International Awards Validate Mini SSD's Innovation and Commercial PotentialSince its market debut, Mini SSD has rapidly gained recognition on the global stage, earning multiple prestigious international awards that highlight its technological innovation, design excellence, and industry impact.International Awards Validate Mini SSD's Innovation and Commercial Potential. Credit:BIWINTIME Best Inventions of 2025Founded in 1923, TIME Magazine is one of the world's most respected media institutions. Its annual Best Inventions list is widely regarded as a global benchmark for breakthrough innovation. Selected from thousands of nominations worldwide by TIME's editorial experts, the award recognizes inventions that demonstrate outstanding innovation, practical value, influence, and future potential.Edison Awards (Finalist)Named after legendary inventor Thomas Edison, the Edison Awards are often referred to as the "Oscars of Innovation." Judged by more than 3,000 global academics, industry leaders, and technology experts, the awards evaluate candidates across four dimensions: concept, value, impact, and achievement. Mini SSD's recognition as a finalist highlights its world-class standing in technology innovation, market potential, and societal impact.CES TWICE Picks AwardsAt CES 2026, widely known as the world's premier technology showcase, Mini SSD stood out among over 4,000 exhibitors and tens of thousands of products to win the TWICE Picks Awards, presented by leading U.S. consumer electronics publication TWICE. The award honors products demonstrating breakthrough innovation, exceptional quality, and strong market potential, and is widely regarded as a mark of excellence in consumer technology.Embedded World "Best-in-Show"The Best-in-Show Award at Embedded World is one of the most respected honors in the embedded systems industry. Evaluated by professional engineers and industry experts, the award recognizes products that excel in performance, reliability, and real-world application value. The recognition underscores Mini SSD's strong potential across embedded and edge computing environments.MWC "Best in Show"At MWC 2026, Mini SSD also received the "Best in Show Award", jointly selected by leading global technology media including TechRadar, Tom's Guide, Android Central, and T3. The recognition highlights Mini SSD's achievements in technological innovation and user experience, further reinforcing its global industry recognition.Mini SSD Receives Multiple Global Industry Awards. Credit:BIWINToday, Mini SSD has rapidly progressed into commercial deployment, gaining adoption among leading device brands. OneXPlayer, GPD, and Waterworld have already integrated Mini SSD into flagship devices including handheld gaming consoles, AI PCs, and smart albums. Early partners report that the adoption of Mini SSD has enhanced user experience while driving measurable market growth and positive brand perception.From global awards to real-world product adoption, Mini SSD is steadily advancing from technology validation to commercial success.Industry Collaboration: Partners Outline the Future of the Mini SSD EcosystemDuring the seminar, industry participants engaged in in-depth discussions around standardization, ecosystem governance, and application deployment, jointly outlining the strategic roadmap for the Mini SSD ecosystem.Industry Collaboration: Partners Outline the Future of the Mini SSD Ecosystem. Credit:BIWINShared Value and Ecosystem GovernanceParticipants agreed on the need to establish a shared-benefit mechanism to accelerate ecosystem development. Plans include forming a dedicated IP management company and governance framework, with clear policies for incentive structures, licensing models, and revenue distribution, ensuring that ecosystem partners share both the value and growth opportunities generated by the technologyDesign-In and Technical CollaborationLeading partners including Longcheer, Emdoor, MINISFORUM, and GMKtec announced plans to initiate design-in programs for upcoming flagship products such as ultra-thin laptops and Mini PCs. Through joint development, design validation, and system optimization, the industry aims to establish a collaborative innovation model across the entire supply chain.Market Expansion and Application StrategyThe seminar also outlined a value-driven market strategy. For PC platforms, participants discussed the possibility of pre-loading popular games and AI models on Mini SSD, creating new value-added distribution channels that enable hardware vendors to unlock additional revenue opportunities.Moving forward, ecosystem partners will work together to accelerate Mini SSD adoption across handheld gaming devices, AI PCs, robotics platforms, and other emerging edge-AI applications.Enabling Edge AI: Mini SSD Unlocks New Possibilities for Next-Generation DevicesWith its ultra-compact footprint (approximately half the size of a coin), PCIe Gen4 flagship-level performance, LGA packaging for enhanced shock resistance, and SIM-card-style expandability, Mini SSD provides a flexible, high-performance storage solution for next-generation intelligent devices.IMini SSD Unlocks New Possibilities for Next-Generation Devices. Credit:BIWINUltra-Thin LaptopsWeighing just 1 gram and offering 40% smaller volume compared with traditional M.2 2230 SSDs, Mini SSD frees valuable internal space within ultra-thin devices, enabling manufacturers to integrate additional components. Its user-replaceable expansion design also allows users to scale storage capacity as AI PCs increasingly require large datasets for local AI model deployment.Intelligent RoboticsFor robotics operating in high-vibration and complex environments, Mini SSD delivers high-speed read/write performance and strong shock resistance, ensuring stable access to vision and environmental data. Its modular architecture enables flexible storage expansion across the robot’s lifecycle, supporting continuous data accumulation for lifelong learning and system evolution.DronesMini SSD’s shock resistance and wide temperature tolerance ensure reliable performance during high-speed flight, rapid maneuvering, and extreme temperature variations. With capacities up to 2TB, it supports stable recording of 4K and 8K aerial video, providing creators with greater storage capacity for professional content capture.Handheld Gaming DevicesIn high-performance handheld gaming consoles from brands such as GPD and OneXPlayer, Mini SSD delivers console-class loading speeds within a footprint only slightly larger than a TF card. With random read performance up to 850K IOPS, it enables near-instant loading for large 3D games, while meeting the strict space constraints of compact gaming devices.Building the Mini SSD Ecosystem Together for the FutureThe Mini SSD Strategic Seminar marked the formal establishment of the Mini SSD ecosystem on the basis of in-depth technical exchange. With a clear roadmap spanning shared-value mechanisms, unified standards, and Design-in collaboration, industry partners are now aligned around accelerating large-scale adoption.Backed by Shenzhen's market accumulation and fast response of the industry chain, BIWIN will continue working closely with ecosystem partners to develop technical standards, expand application scenarios, and build a thriving Mini SSD industry platform.

As the global satellite communications industry rapidly advances toward a multi-orbit, multi-band architecture, the flexibility and scalability of ground terminal equipment have become core concerns for operators. Ultraband Technologies announced it will showcase its 16×16 Ku-band phased array module at the Taiwan Space Pavilion during Satellite 2026. Through a modular design philosophy, the company delivers more flexible and scalable terminal solutions for multi-orbit, multi-band satellite communications.Against the backdrop of multi-orbit, multi-frequency satellite networks gradually taking shape, ground terminal equipment must simultaneously support satellite systems operating across different orbits and frequency bands. However, when implementing multi-orbit, multi-frequency architectures, operators face the common challenge of rapidly escalating costs associated with mass production and large-scale deployment for terminals centered around phased array antennas. To address this need, Ultraband Technologies proposes a predictable and consistent ground terminal evolution strategy. This approach assists global satellite operators managing multiple satellite constellations in maintaining technological platform consistency while effectively addressing cost and performance challenges during mass production and deployment phases.Modular Stitching Architecture: Supports gradual expansion from UAV BLOS (Beyond Line of Sight) applications to large ground terminal equipment. Operators can enhance system capabilities through modular expansion, adapting to different orbital or frequency strategies and mission requirements without replacing the entire platform.Multi-Orbit, Multi-Band Switching Design: Collaborating with industry partners to enable a single terminal device to support multi-orbit and multi-band switching capabilities. This allows equipment to operate flexibly across different satellite orbits and frequency band systems, further reducing terminal SKU counts and lowering overall operational costs.Gallium Nitride (GaN) Technology: Integrating proprietary GaN design and process capabilities, this technology enhances signal output power while maintaining low power consumption and superior thermal performance. This ensures high stability and reliability for equipment operating in prolonged and all-weather environments.Ultraband Technologies'newly launched 16×16 Ku-band phased array module serves as a Proof of Concept (POC) and early testing platform, open for evaluation and validation by satellite operator technical teams and research institutions. Through real-world application testing, it assists industry partners in verifying the feasibility of antenna architecture design and field deployment processes prior to formal commercial deployment.For multi-orbit, multi-band operators, this signifies that phased array ground equipment is evolving from a high-performance technology demonstration into deployable, repeatably scalable, and operationally ready infrastructure, laying the foundation for the future large-scale development of satellite communication networks.In the drive toward seamless system implementation, Ultraband Technologies is collaborating with a strategic network of domestic and international partners - including Ubiqconn Technology - to leverage extensive expertise in ruggedized system integration and manufacturing.By introducing a cutting-edge Fast Calibration Mechanism, the collaboration effectively addresses the industry's mass production bottlenecks. This innovation significantly shortens calibration cycles for phased array systems during the manufacturing stage, ensuring superior batch consistency and drastically reducing system tuning costs for large-scale deployments.Ultraband Technologies will showcase the aforementioned phased array modules and related technological achievements during the Satellite 2026 exhibition, engaging in collaboration and exchange with global satellite operators, system integrators, and industry partners.Satellite 2026 Exhibition InformationExhibition Dates: March 23–26, 2026Venue: Walter E. Washington Convention CenterBooth Number: Taiwan Space Pavillion #2747Click here for more information.Ultraband

As global AI computing platforms continue to evolve toward higher density, increased power demands, and rack-scale integration, Chenbro unveiled its latest AI server chassis solutions at NVIDIA GTC 2026. The showcase features the NVIDIA MGX 1U Compute Tray for NVIDIA Vera Rubin NVL72 systems, along with a comprehensive product portfolio based on NVIDIA MGX architecture, demonstrating Chenbros integrated capabilities spanning system design, thermal optimization, and rack-level deployment.Building on its close integrations with the NVIDIA ecosystem, Chenbro presented multiple server chassis solutions designed in alignment with NVIDIA reference architectures and the NVIDIA MGX architecture. These solutions address high-density computing, liquid cooling technologies, and enterprise-grade deployment requirements, enabling cloud service providers(CSPs) and enterprise data centers to accelerate AI infrastructure deployment.NVIDIA MGX 1U Compute Tray for NVIDIA Vera Rubin NVL72- A New Standard for High-Density AI ComputingAs next-generation AI training and inference workloads continue to scale rapidly, data centers face increasing demands for space efficiency and advanced thermal design. At NVIDIA GTC 2026, Chenbro highlights the NVIDIA MGX 1U Compute Tray for Vera Rubin NVL72, engineered with optimized mechanical design and airflow management to deliver high-density computing within a 1U footprint. The solution is tailored for large-scale model training and high-performance inference applications.Chenbro also showcases:NVIDIA MGX 1U for NVIDIA GB200 NVL4, supporting high-performance inference and flexible deployment.NVIDIA MGX 2U Vera Server Chassis, offering enhanced scalability and enterprise-level integration options.Together, these solutions demonstrate Chenbro's system integration expertise under the NVIDIA MGX architecture framework.NVIDIA MGX Rack- Advancing from System Integration to Rack-Scale DeploymentAs AI infrastructure expands from individual systems to rack-scale environments, Chenbro highlights its integration capabilities based on NVIDIA MGX architecture. The company leverages its strengths in mechanical design and mass production to support the deployment of high-density AI computing environments in modern data centers.By closely aligning with NVIDIA reference architectures and the broader NVIDIA MGX ecosystem, Chenbro enables customers to extend from system-level builds to rack-level integration, accelerating scalable AI infrastructure deployment.NVIDIA MGX 6U Liquid-Cooled Server Chassis- Liquid Cooling for High-Power AI PlatformsIn response to power consumption and energy efficiency requirements for AI platforms, Chenbro presents the NVIDIA MGX 6U Liquid-Cooled Server Chassis. Designed to support high-power liquid-cooled deployment architectures, the solution enhances thermal efficiency and improves data center power usage effectiveness (PUE) through optimized mechanical and fluid management design, ensuring stable operations for next-generation AI data centers.Additional showcased solutions include:NVIDIA MGX 4U Air-Cooled Server Chassis and NVIDIA MGX 2U Short-Depth Server Chassis.These offerings provide flexible deployment options for enterprise server rooms and diverse application scenarios.Strengthening Global Footprint and AI Infrastructure CapabilitiesChenbro CEO Corona Chen stated that as AI applications continue to expand, market demand for high-density computing, liquid cooling, and rack-scale integration capabilities is rapidly increasing. Chenbro will continue deepening its collaboration within the ecosystem, strengthening its integrated R&D and manufacturing capabilities to help customers accelerate the adoption of next-generation AI platforms.                                                      With its global manufacturing footprint and localized service capabilities, Chenbro is committed to supporting diverse deployment needs across AI, HPC, and data center applications—driving sustained operational growth and industry competitiveness.At GTC 2026, Chenbro not only presents its product portfolio but is supporting NVIDIA AI platforms from system integration and thermal optimization to rack-scale deployment, enabling scalable AI infrastructure development worldwide.Chenbro supports diverse AI, HPC, and data center deployment needs. Credit: Chenbro