On the opening day of COMPUTEX 2026, AIC Inc. hosted a high-level strategic panel session at its booth, focusing on overcoming the "memory wall" challenge. Industry giants and key strategic partners, including NVIDIA and VAST Data, joined AIC for a presentation on their latest platforms designed to eliminate bottlenecks in Large Language Model (LLM) inference and intensive AI workloads, marking a critical evolution in active AI storage driven by Agentic AI in 2026.In his opening remarks, Michael Liang, CEO and President of AIC, outlined the new challenges facing AI infrastructure as AI applications enter the "Long Context" era. The transition to long-context and Agentic AI has completely shifted the primary AI infrastructure bottleneck from raw computational speed to massive data movement and memory bandwidth constraints—a hurdle commonly known as the "Memory Wall."Liang emphasized that the shift toward autonomous AI agents executing task decomposition and multi-step APIs is fundamentally transforming data center demands and reshaping underlying AI infrastructure. Consequently, AIC is actively collaborating with NVIDIA and VAST Data to develop advanced, AI-native storage solutions. By integrating the NVIDIA Vera BlueField-4 STX Storage Processor into its hardware platforms, AIC is building the essential infrastructure required to eliminate bottlenecks and accelerate workloads for Agentic AI applications.NVIDIA Ecosystem Scales Agentic AI Adoption WorldwideJason Hardy, NVIDIA's Vice President of Storage Technology, highlighted the significance of "Agentic Inferencing", a key theme from the NVIDIA GTC Taipei keynote during COMPUTEX 2026. Agentic AI requires more than faster compute; it demands fast, secure access to context memory so agents can reason across long sessions, large datasets, and complex workflows.NVIDIA Vera BlueField-4 STX addresses this paradigm shift. It enables a new class of AI-native storage infrastructure for context memory, built with Vera-based BlueField-4, NVIDIA Spectrum-X Ethernet, NVIDIA DOCA, NVIDIA Dynamo, and NVIDIA AI Enterprise. This foundation provides NVIDIA's storage partners, such as AIC, with the essential building blocks to keep agent context and inference data close to the compute path, significantly improving throughput, responsiveness, and infrastructure efficiency.NVIDIA is actively building a robust partner ecosystem around the NVIDIA Vera BlueField-4 STX architecture, spanning storage, systems, cloud infrastructure, and security sectors. Key partners like AIC are collaborating closely with NVIDIA to integrate, validate, and bring this next-generation infrastructure to market. Hardy emphasized that these close alliances will help customers optimize resource utilization, reduce costs, accelerate response times, and enhance security during large-scale deployments, thereby ushering in the era of Agentic Inferencing.VAST Data and AIC Hard-Soft Integration Optimizes AI InfrastructureEchoing the new design of NVIDIA Vera BlueField-4 STX, VAST Data CTO Andy Pernsteiner emphasized that Agentic AI requires sophisticated mechanisms for managing and optimizing massive-scale KV caching to persistent memory. This avoids redundant, expensive prefill computations across multi-turn, long-context sessions, while providing new storage platforms that support confidential computing and data protection for highly sensitive information. VAST Data integrates seamlessly with NVIDIA's BlueField-4 DPU architectures and Dynamo routing frameworks to offload, share, and reuse KV cache context across wide GPU clusters.The strategic hardware-software partnership between VAST Data and AIC pairs AIC's advanced server hardware with VAST's software intelligence to build next-generation AI infrastructure and context memory storage platforms. Integrating NVIDIA Context Memory Storage (CMX) platform, featuring the NVIDIA Vera BlueField-4 STX storage processor, effectively resolves GPU KV cache bottlenecks. By utilizing fast NVMe arrays as a shared, high-bandwidth context tier, the solution significantly increases tokens-per-second throughput and energy efficiency for long-context, multi-turn AI inferencing.AIC Embraces NVIDIA Vera BlueField-4 STX for Agentic AIAs Liang stated in a post-event interview with DIGITIMES, the company has successfully built its storage server business since 2014. By continuously reinvesting 15% of its annual revenue every year into R&D and early-stage development of new architectures, AIC has positioned itself as a key player in developing next-generation, Agentic AI-native storage infrastructure.Today, AIC is established as a Solution Advisor in the NVIDIA Partner Network (NPN). AIC also is building upon a strategic partnership with VAST Data that began seven and a half years ago. To meet the surging demand for AI data centers, AIC's strategic expansion in Yangmei, Taiwan, and Haiphong, Vietnam, directly targets the skyrocketing global demand for artificial intelligence data centers. These state-of-the-art manufacturing footprints allow the company to scale production of AI servers and high-density storage while seamlessly integrating computing, networking, and security into unified infrastructure platforms.The new facilities anchor AIC's global supply chain and position the company to meet the intense deployment needs of cloud service providers and enterprise customers. This empowers customers to maintain a competitive lead and achieve greater success amidst the AI wave.

The megatrend in electronic design today is end-to-end collaboration across ICs, packaging, PCBs, systems, data centers, and physical applications, with rapidly evolving artificial intelligence playing an increasingly critical role.In early June, Graser Technology held its annual technology forum, Graser TECHTALKS 2026, under the theme "AI in Sync: Intelligent Design, Accelerated Manufacturing." The event focused on how AI connects design, analysis, and manufacturing workflows. It brought together industry speakers, in-house engineering experts, and customer representatives to share professional insights and real-world experience, outlining a new paradigm for electronic design workflows and industrial applications in the AI era.In her opening remarks, Graser Chairwoman Lillian Pan said the company has, for more than 30 years, upheld the principles of fast response, professional service, and long-term partnership, helping customers turn ideas into products faster. She added that Graser will continue promoting the leverage of AI across engineering workflows, introducing advanced design tools, and supporting Taiwan's semiconductor and electronics industries in remaining globally competitive.AI as a Design Workflow CollaboratorIn the first keynote, "Paradigm Shift of System Design in the AI Era," Michael Shih, Corporate Vice President for APAC and Japan at Cadence, said electronic design is facing a new level of complexity as Moore's Law becomes harder to sustain and the cost of advanced process technologies and system integration continues to rise.He noted that the challenge is no longer limited to designing a single chip. Instead, engineering teams must increasingly solve complex issues across chips, advanced packaging, PCBs, system-level design, and multiple physical domains. Against this backdrop, Cadence has been expanding its focus from IC design into packaging, PCB design, multiphysics simulation, data centers, and system analysis, evolving from a traditional EDA tool provider into an Intelligent System Design platform company.Shih explained that Cadence's Intelligent System Design platform brings together AI, EDA and IP, system design and analysis, and computational software. This enables engineering teams to perform simulation, analysis, optimization, and design verification at the system level. Within this framework, Cadence is pursuing AI in two directions: Design for AI, which helps customers build AI infrastructure, and AI for Design, which embeds AI directly into design solutions. In other words, AI is not only an application enabled by advanced ICs and systems; it is also becoming a core collaborative capability within the electronic design process.A major part of this shift is the introduction of agentic AI into design workflows. Shih said Cadence is bringing AI agents into front-end design and verification, digital implementation, and custom and analog design processes.These AI agents can help engineers understand design goals, break down tasks, execute workflows, and accelerate iterative design cycles. Their value goes beyond labor savings: by automating repetitive and time-consuming work, AI agents allow design teams to explore feasible options faster, shorten development cycles, and reduce the time and cost pressures created by rising complexity of designs.Shih noted that, for example, many companies must complete large numbers of board designs every year, involving repetitive yet expertise-intensive tasks such as placement, routing, layout, and design checks. By introducing AI into these workflows, engineers can spend more time on system architecture, reliability, and innovation. This suggests that design automation in the AI era is moving beyond point-tool acceleration toward broader efficiency gains across ICs, packaging, PCBs, and system-level simulation.AI Deployment Through System IntegrationFocusing on system integration design trends in the AI era, Eric Kao, Business Development Director at Giga Computing, shared his perspective from the data center infrastructure side. He noted that as enterprises adopt AI agents and generative AI applications, inference workloads are growing rapidly, pushing data center architectures originally optimized for AI training to shift.This shift is also redefining the role of the CPU. Because AI agent workflows involve task decomposition, step-by-step planning, API calls, tool invocations, and other logic-heavy and I/O-intensive operations, the CPU is no longer just a supporting component next to GPUs or accelerators. Instead, it is becoming the control and orchestration hub inside the AI data center.Kao pointed out that future AI infrastructure will move toward more refined heterogeneous computing configurations. Effectively managing different platforms and resources—and matching the right hardware to the right models and workloads—will become a critical system design challenge.Giga Computing's own technology roadmap also reflects this transition. According to Kao, the company has expanded from server motherboards and system development into HPC, OCP, GPU servers, liquid cooling, heterogeneous computing platforms, and broader AI infrastructure services. This shows that competition in AI data centers is shifting from standalone server specifications to integrated capabilities across racks, cooling, networking, software, POD design, and system-level simulation.Po-Ting Lin, Professor in the Department of Mechanical Engineering and Director of the Center for Intelligent Robotics (CIR) at National Taiwan University of Science and Technology (NTUST), approached AI from the perspective of physical system applications. He shared his team's experience applying AI to obstacle-avoiding path planning for robotics.Lin explained that when a robot encounters nearby people or obstacles during operation, it must quickly determine a safe trajectory to avoid collisions. Traditional optimization methods can be used to search for safe paths, but they often require significant computation time. By incorporating AI models, the system has the potential to greatly shorten response time.Lin emphasized that robot obstacle avoidance is not about taking the longest possible detour. The goal is to find a path that avoids obstacles just enough while maintaining task efficiency. NTUST's robotics research covers human-robot collaborative robotic arms, UAV inspection, and dual-arm robotic systems, with a common focus on balancing safety and operational efficiency.Through the insights shared by these two speakers, it is evident that bringing AI into real-world applications depends not only on a single chip or algorithm but also on the integration of computing, software, sensing, simulation, and physical systems.Intelligent Tools and Simulation Integration Across the Design FlowThe afternoon sessions of Graser TECHTALKS 2026 focused on two major tracks: electronic system design automation and multiphysics simulation. Graser's engineering team highlighted the latest advances in Cadence Allegro/OrCAD X 25.1 and Allegro X AI, demonstrating how automation and AI-assisted design can improve PCB development workflows.The program also featured technical experts from AIC, Supermicro, and Cadence, who shared practical insights into power integrity, electrothermal co-simulation, AI server system design, and multiphysics optimization, spanning packaging to system-level design, using Cadence Sigrity, Clarity, Celsius 3D, Sigrity HPC, and Aurora.Graser also presented updates to its in-house software portfolio, including GraserWARE, GIMS, and CAMPro, addressing requirements such as circuit reliability checks, component and BOM management, and manufacturing data validation.Building on features introduced last year, the company added several practical tools to GraserWARE MSAPack, including simulation schedule management, stackup format conversion, S-parameter port-naming optimization, temperature-dependent material parameter fitting, and automatic Power Tree generation. These capabilities help streamline SI/PI simulation workflows while improving analysis efficiency and data consistency.The key takeaway from Graser TECHTALKS 2026 is that in the AI era, design competitiveness goes beyond upgrading individual tools—it depends on how effectively organizations can synchronize design, analysis, verification, and manufacturing data to enable faster, more agile system-level development.

In the past, due to high equipment costs and bulky sizes, thermal imaging technology was predominantly confined to specialized fields such as security surveillance, firefighting and rescue, military defense, and high-end industrial inspection. However, driven by declining sensor costs, maturing AI algorithms, and rapid advancements in module miniaturization, thermal imaging applications have expanded from specialized niches to a broader range of commercial and consumer markets. These include predictive maintenance, EV battery monitoring, and AI-automated inspection.The global thermal imaging market is projected to grow from USD 9.21 billion in 2026 to USD 14.51 billion by 2030, representing a compound annual growth rate (CAGR) of approximately 12%. To help Taiwanese industries capture this massive market potential, Suntek Global has partnered with Teledyne FLIR to establish a local thermal imaging ecosystem. Beyond offering a comprehensive product portfolio, Suntek delivers multifaceted technical support services, providing an all-in-one solution that accelerates deployment for local enterprises.Jason Ray, CEO & Managing Partner of Suntek Global, stated that Teledyne FLIR is the undisputed global leader in the commercial thermal imaging market, offering an exceptional range of sensors, focal plane arrays, and thermal calibration solutions recognized worldwide for their quality and diversity. Suntek focuses on localized integration and engineering deployment services, covering custom carrier boards, firmware development, AI model integration, IP-rated enclosure design, regulatory certification, module QA calibration, and localized technical support. The goal of this joint ecosystem is to provide Taiwanese industries with ready-to-mass-produce thermal imaging solutions. Backed by Teledyne FLIR's OEM resources, Suntek assists clients through reference design, thermal module integration, firmware development, and mass production deployment. Suntek's mission is to ensure that Taiwanese device makers do not have to become thermal experts to build thermal-enabled products — they can leverage Suntek's expertise, integrate, and go to market.AI and Cost Reduction Drive Thermal Imaging into Mass MarketsBenefiting from the continuous decline in thermal module costs, a wide range of new application scenarios has emerged. These include smart buildings, energy management, home security, robotics, automation equipment, and consumer electronics, driving thermal imaging from niche professional domains into mass markets. Notably, unlike traditional night-vision cameras, thermal imaging modules detect the thermal contours emitted by humans or objects. This uniquely satisfies the dual demand for advanced sensing and privacy protection in various settings. Consequently, these solutions are being widely adopted in smart buildings, long-term care facilities, public spaces, and smart city developments.Taiwan stands as a global hub for semiconductors and high-tech manufacturing. While many local OEMs and ODMs possess robust hardware manufacturing capabilities, integrating thermal imaging modules into products inevitably presents hurdles such as radiometric calibration, FFC (Flat Field Correction), and ISP tuning—tasks that typically require substantial time and talent investment. To address this, Suntek Global aims to lower the barrier to entry through localized support services, allowing companies to integrate thermal imaging into their product design cycles more efficiently. From knowledge sharing and training to reference designs, Suntek is fully committed to building Taiwan's thermal imaging ecosystem.A prime example is AiForce, a company specializing in thermal imaging sensors, optical components, and AI vision training integration, dedicated to creating high-performance, multi-scenario intelligent vision systems. With full technical backing from Suntek Global, AiForce successfully compressed its product development timeline, enabling it to launch its new solutions to the market in the shortest time possible.Ethen Zhong, CEO of AiForce, noted, "Teledyne FLIR's thermal imaging components are undoubtedly top-tier global products. However, because of this, it is challenging for the original manufacturer to provide immediate, localized support to every regional user, making a dependable distributor indispensable. Although we are highly proficient in AI and thermal imaging component integration, we still encountered various calibration and parameter adjustment challenges during development. This is where Suntek Global played a critical role. Suntek Global is no longer just a traditional distributor; they act as a vital bridge connecting original manufacturer technology, local engineering resources, and customer needs. By helping us resolve module integration, image tuning, software-hardware validation, and production introduction issues, they enabled AiForce to focus entirely on application innovation and accelerate our commercialization timeline."From Technical Support to Talent Cultivation: Building Taiwan's Thermal EcosystemAs global demand for thermal imaging surges, product competitiveness will no longer rely solely on hardware specifications. Instead, it will hinge on the integration capabilities across sensors, algorithms, AI inference, optics, and system platforms. To this end, Suntek Global is mapping out its ecosystem across four core pillars to help Taiwanese industries capture this massive market potential.Jason Ray pointed out that the first pillar is knowledge promotion, which begins with the launch of the "Thermal Tech" column to provide in-depth technical analysis and case studies. The second is talent training, offering courses that cover thermal integration, ISP tuning, and AI model deployment. The third is reference design and demonstration platforms to accelerate client development processes. Finally, the fourth pillar focuses on vertical industry application showcases, ensuring partners stay informed of the latest trends and practical use cases.From sensors and modules to AI vision integration, the thermal imaging market is rapidly entering a new growth cycle.  The thermal revolution is no longer on the horizon — it is already reshaping Taiwan's manufacturing landscape. Suntek Global, in partnership with Teledyne FLIR, intends to be the connective tissue that makes Taiwan's leadership in this space inevitable.Teledyne FLIR Boson thermal camera module. Credit: Suntek GlobalThermal imaging application for industrial monitoring. Credit: Suntek Global

The integration of artificial intelligence into the semiconductor industry is no longer merely a competitive advantage - it has become a core driver of modern engineering.The Industrial Development Administration (IDA) under Taiwan's Ministry of Economic Affairs is committed to building a comprehensive training and support ecosystem for international talents in Taiwan. The initiative covers key semiconductor technologies, cross-cultural communication, and local integration, while also leveraging online learning resources to connect talent across the Asia-Pacific region.Designed to help outstanding international students and professionals transition seamlessly into Taiwan's semiconductor industry, the program provides early exposure to the local industrial ecosystem, along with mentorship and tailored support services that strengthen long-term adaptation, professional growth, and career development in Taiwan. Through these efforts, Taiwan aims to cultivate a more inclusive and globally connected environment for international talent, further enhancing the global competitiveness of its semiconductor industry.Through specialized talent development programs in Taiwan, Wifal Inola from Indonesia and Earon John Mendoza from the Philippines are transforming their professional capabilities and expanding their roles in the global semiconductor ecosystem.By bridging the gap between theoretical AI knowledge and its practical application in high-tech manufacturing and research and development, these programs are cultivating a new generation of interdisciplinary talent prepared for the future of advanced technology industries.Advancing technical depth and specialized applicationThe training provided in Taiwan offers a significant shift in both technical depth and industrial focus compared with the educational experiences available in the engineers' home countries.Wifal Inola, a master's student in the Department of Semiconductor Technology at National Yang Ming Chiao Tung University (NYCU), observed that while AI training in Indonesia often focuses on supporting the digital economy through e-commerce and financial technology, Taiwan's curriculum is deeply rooted in advanced semiconductor applications and system integration. This includes specialized fields such as smart manufacturing, medical technology, and robotics.Similarly, Earon John Mendoza, a QW1612 Assistant Engineer at ASE and a post-baccalaureate student at I-Shou University, emphasized that AI-related training in Taiwan is far more structured and technically rigorous. He noted a clear contrast with his previous educational experience in the Philippines, where the focus was often placed on final output and task completion under pressure.In Taiwan, however, the emphasis is placed on understanding every step involved in building AI models - ensuring engineers understand not only how a process works, but why each stage is necessary. This foundational approach is especially critical in semiconductor manufacturing, where skipping steps in troubleshooting or equipment maintenance can lead to serious systemic failures.Enhancing efficiency through AI tools in R&D and manufacturingBoth engineers have successfully integrated AI tools into their daily workflows, reducing manual workloads and allowing greater focus on high-value technical decision-making.Wifal Inola, who secured both an internship and a future full-time position at Micron as a Process Engineer in Diffusion Process, uses AI to streamline the demanding research process required in semiconductor studies. He applies AI tools to significantly reduce the time spent on literature reviews, allowing him to quickly understand unfamiliar research topics.He also leverages AI for technical analysis tasks such as X-ray Photoelectron Spectroscopy (XPS) peak division and chemical bond identification. By delegating repetitive analytical work to AI, Wifal is able to focus more on designing better experiments and optimizing process parameters.In the industrial sector, Earon applies AI knowledge within ASE's preventive maintenance operations to ensure production lines remain stable and efficient. His work focuses on hardware-related systems such as sensors, controllers, and equipment monitoring.In an environment where every minute of machine downtime translates into significant production losses, AI becomes a critical tool for operational efficiency. Earon uses AI-assisted inspection systems and computer-based monitoring tools to track equipment failures and identify non-good (NG) parts with high precision.His background in mechatronics, combined with AI training, allows him to better understand how different hardware components interact - helping him identify faster and more effective solutions while improving productivity beyond standard performance targets.Localized AI development beyond manufacturingBeyond hardware and manufacturing applications, both Taiwan and other countries are also developing localized AI models to better serve their own linguistic and cultural environments.Wifal noted that Taiwan has developed TAIDE (Trustworthy AI Dialogue Engine), which focuses on traditional Chinese language applications and local cultural context. At the same time, Indonesia has introduced Sahabat AI, a model designed specifically for Bahasa Indonesia and regional dialects.This development highlights that AI training is not solely about meeting global technical standards - it is also about understanding how technology can be adapted to serve local populations more effectively.The future: from manual labor to strategic decision-makingBoth engineers believe AI will fundamentally reshape the semiconductor industry over the next five to ten years. Wifal believes AI will eventually take over most repetitive and manual tasks, shifting the role of engineers toward high-level decision-making based on AI-assisted analysis and predictive systems.This transformation will require a new type of professional - one who combines deep semiconductor expertise with strong capabilities in data analysis and AI fundamentals.Earon shares a similar view, expecting AI to become a key solution for reducing time-consuming and physically demanding tasks that engineers currently perform manually. This will allow professionals to focus more on solving complex technical challenges and driving innovation.As AI continues to automate repetitive work and redefine traditional engineering roles, the experience gained in Taiwan ensures these professionals remain at the forefront of industry transformation.Whether they return to their home countries or continue their careers at world-class companies such as Micron and ASE, they carry with them the technical depth, operational efficiency, and strategic mindset needed to lead AI-integrated engineering teams.Ultimately, this talent exchange creates a true win-win scenario: Southeast Asian engineers gain access to world-class career opportunities, while the global semiconductor industry benefits from a more resilient, technically sophisticated, and future-ready workforce.A new era of global synergyThe stories of Wifal Inola and Earon John Mendoza reflect a broader shift in the global semiconductor landscape - one where talent mobility and specialized AI training are becoming key drivers of innovation.By opening its doors to engineers from Indonesia and the Philippines, Taiwan is doing more than addressing talent shortages. It is fostering an advanced ecosystem of professionals who are fluent in both semiconductor engineering and artificial intelligence.For the engineers themselves, the journey is transformative - turning them into interdisciplinary leaders capable of navigating the growing complexity of modern manufacturing, process optimization, and R&D.As the semiconductor industry becomes increasingly AI-driven, these talent development programs are ensuring that the next generation of Southeast Asian engineers is not simply adapting to change - they are prepared to lead it.   

LITEON Technology is actively capitalizing on the AI boom by restructuring its data center infrastructure around megawatt-scale power, rack and liquid cooling. The company has developed high-density 800 VDC power architectures and advanced cooling systems to support massive, next-generation AI accelerators.At COMPUTEX 2026, LITEON hosted the "AI Summit Panel" on the opening day, themed "Powering the AI-driven Future: Scaling AI Across Architecture, Systems, and Infrastructure." Moderated by Colley Hwang, Chairman of DIGITIMES, the Panel brought together industry leaders from NVIDIA, Infineon and GIGABYTE to examine how AI is reshaping the technology landscape. The discussion highlighted how rapid AI adoption is driving structural changes brought across the ecosystem, particularly in data center infrastructure and power demand. As AI scales, power availability and efficiency are becoming central considerations, with AI factories emerging as a defining paradigm in the next phase of industry development.Hwang opened by noting that AI has become a key driver of global productivity growth and is expected to generate trillions of dollars in economic value. As a result, infrastructure and deployment models are being fundamentally reshaped, with data centers rapidly evolving into AI factories. In this transition, the availability and reliability of energy have emerged as critical enablers of large-scale deployment, elevating power and cooling from supporting functions to core elements of system architecture.More than just chips, AI revolution requires huge energyNVIDIA Senior Director, HPC and AI Hyperscale infrastructure Solutions, Dion Harris spoke first. He referenced NVIDIA CEO Jensen Huang's speech in the NVIDIA GTC Taipei held in the previous day, addressing the "five-layer cake" analogy to illustrate AI from being just a software tool into a vast, vertically integrated industrial system. At the very bottom of the cake is "energy". Before the AI can showcase real-time intelligence, the systems fundamentally rely on efficient and stable power supply."Building AI factories goes beyond compute-you have to power them, cool them, and run them reliably to truly monetize AI," said Dion Harris at NVIDIA. "Infrastructure and energy play a critical role in determining token cost and performance per watt. As Jensen Huang's‘five-layer cake’ illustrates, without strong foundational layers, higher-level innovation cannot scale. This is why ecosystem collaboration is essential."Harris further mentioned NVIDIA DSX platform as a blueprint for AI factory development, defining how next-generation infrastructure is designed, built and operated. Under this framework, AI factories are evolving to support rapidly growing token demand driven by today's AI applications. He emphasized that achieving optimal token cost and performance per watt ultimately relies on strong ecosystem collaboration, noting that partners like LITEON play a critical role in delivering the power, cooling, and operational capabilities required to support frontier and open-source AI workloads at scale.Dr. Sergio Rossi, Vice President of Application Marketing at AI and Power division of Infineon Technologies, noted that AI adoption accelerates, the key competitive bottleneck is shifting from compute to energy, especially in regions such as Europe where electricity availability has become increasingly constrained. That's the reasons why LITEON, NVIDIA, GIGABYTE and Infineon engage in collaboration to have several steps forward to make sure the design is using the right technology, architecture and solutions for securing AI development to use energy in the most efficient manner."Just improving energy efficiency by a few percentage points can translate into saving power at the scale of an entire city, highlighting how critical efficiency has become in AI data center development," said Dr. Serio Rossi.Rossi further highlighted that the AI ecosystem is facing three key complexities, including acceleration, with development cycles shortening from 30 months to 6 months; increasing architectural complexity driven by the adoption of 800 VDC in next-generation data centers; and rapid innovation in power technologies such as SiCs and GaNs to enhance energy efficiency. To address these challenges, he emphasized the importance of close ecosystem collaboration, where infrastructure stakeholders work together to anticipate future demand and align on technology roadmaps with partners such as Infineon at an early stage. This enables faster development cycles and accelerates innovation in power and infrastructure solutions for evolving AI requirements.Leo Wang, EMEA regional Product & Marketing Lead at Giga Computing, talked about how the company collaborates with industry leaders to advance AI performance through hardware innovation and liquid-cooled server solutions optimized for AI workloads. He noted that as AI infrastructure evolves, the focus is extending beyond server performance to broader data center challenges, particularly in power and thermal management.Wang added that maximizing the value generated from available energy is becoming a key industry priority, making close ecosystem collaboration essential. Leveraging Taiwan's strong supply chain and close engagement with global partners, this collaboration is helping to establish scalable deployment models and best practices for global AI adoption."We are now dealing with increasingly heterogeneous machines, from training systems to different AI applications, so power and space both need to be carefully managed, as all of these resources are limited," said Leo Wang at Giga Computing, "Everything comes down to how we build a proper data center and a proper environment to serve the server in a consistent way. It is no longer just about building servers, but about optimizing at the data center level to ensure power is allocated and utilized in an effective and efficient way."From Power to Infrastructure: Enabling the Next Phase of AI ScalingIn response to the energy and infrastructure challenges highlighted in the panel, LITEON showcased its integrated capabilities across power, mechanical, and thermal systems to support next-generation AI infrastructure at COMPUTEX 2026. LITEON debuts 800 VDC liquid-cooled power rack, alongside a comprehensive portfolio of solutions, including the 110kW power shelf for NVIDIA Vera Rubin NVL72 platforms and the 280kW in-rack CDU. Together, these offerings address the growing demand for high-density and energy-efficient AI infrastructure, demonstrating LITEON's ability to integrate power delivery, thermal management, and system design to support scalable AI deployment, and reinforcing its role as a key enabler in the evolution toward megawatt-scale AI data centers.

Clientron, a premier global provider of smart endpoint solutions, today announced a strategic partnership with Parallels, a global leader in cross-platform and virtualization solutions. Through this collaboration, Clientron will bring Parallels Workspace solutions including Parallels RAS (Remote Application Server), Parallels Browser Isolation and Parallels DaaS to organizations across Southeast Asia, offering a simple, flexible, and secure alternative for businesses looking to modernize digital workspaces and reduce IT cost and complexity.In today's hybrid work era, IT teams across Southeast Asia are increasingly burdened by "Legacy VDI Platforms" and are increasingly looking for ways to simplify digital workspace management, reduce operational overhead, and improve cost efficiency. Parallels Workspace solutions address these challenges with streamlined deployment, flexible infrastructure support, and cost-effective application and desktop delivery across hybrid and multi-cloud environments."Businesses want to stop using old and costly systems," said Vivienne Weng, Vice President at Clientron. "Parallels is a great way to reduce the work for IT teams. With Clientron's high-performance hardware, we help IT managers in Asia move away from expensive setups to a smarter and easier way to manage their workspaces.""Many organizations across Southeast Asia are currently seeking solutions to modernize their digital workspace strategies," said Asif Khan, Country Sales Manager at Parallels. "As a strong mid-market enabler, Parallels helps organizations simplify application and desktop delivery with secure, flexible solutions designed for hybrid and multi-cloud environments - backed by competitive and cost-effective pricing. Together with Clientron, we're expanding access to these modern workspace solutions across the region."Key Advantages of Clientron x Parallels:1. Cost-Effective Seamless Access: Smoothly run critical apps on Mac, Windows, or Linux platforms while cutting Total Cost of Ownership (TCO); 2. Deployment Without Complexity: Parallels Workspace Solutions can be deployed in hours with minimal specialized training; 3. Ultimate IT Flexibility: Support workloads across hybrid, multi-cloud, and on-premises environments, protecting enterprises from vendor lock-in; 4. Stronger Security: Modern encryption and login technology to keep your data safe with high-class protection while staying fast and smooth; and 5. Localized Southeast Asia Support: Dedicated focus on key Southeast Asian markets including Malaysia, Thailand, the Philippines, Taiwan, Vietnam, and Indonesia, helping customers and partners access localized expertise, deployment support, and faster regional engagement.Moving forward, Clientron will continue to expand its footprint in Asia, working closely with channel partners to promote Parallels solutions and help enterprises build 'Smart Secure Workspaces' for the future.Deliver secure apps and desktops via Parallels RAS with simple VDI and seamless remote access.Credit: Clientron

Rugged edge computing brand – Cincoze will participate in Automate 2026 in Chicago, USA, from June 22–25 (Booth #861). Under the theme of "Empowering Smart Automation with Edge AI," Cincoze will present its complete range of Edge AI computing and automation solutions across four dedicated zones, including DIN-Rail Computers, Industrial Panel PCs and Monitors, Rugged Embedded Computers, and GPU Embedded Computers.DIN-Rail Computers Zone: The Optimal Computing Platform for Smart Manufacturing and Machine VisionThe Machine Computing — MAGNET product line is engineered for machine vision and smart manufacturing. The flagship MD-3000 Series, winner of the 2026 iF Design Award, will be on display, showing its powerful computing performance, flexible expandability, compact form factor, and DIN-rail installation. In addition to wide temperature and wide voltage support and EN 61000-6-2/6-4 industrial environment EMC compliance, the MD-3000 Series is powered by Intel Core desktop-class CPUs and supports up to 96 GB of DDR5 memory and dual NVMe SSDs, effortlessly handling the high-speed image processing and large-scale data storage demands of machine vision. The modular scalable architecture supports 2-, 4-, or 6-slot expansion decks with a variety of I/O interfaces, PoE modules, M.2 slots, and storage options to meet diverse integration needs. At just 150 mm in height, the compact chassis mounts on DIN rails for flexible deployment in production-line control cabinets, overcoming the challenges of space-constrained installations.Industrial Panel PCs and Monitors Zone: Complete HMI Solutions for Modern Factory ApplicationsThe Display Computing — CRYSTAL product line offers nearly 250 configurations of industrial panel PCs and monitors, covering a wide range of computing performance levels, screen sizes, display ratios, and touch technologies for harsh indoor industrial environments, high-brightness outdoor settings, and open-frame equipment integration. The entire series supports a wide temperature range and a wide voltage input, and is waterproof and dustproof, making it the go-to choice for HMI solutions in demanding industrial environments. The newly launched CV-200 Series features a ≤ 3 mm ultra-slim bezel design that combines performance with aesthetics. Its 178° wide viewing angle, Full HD resolution, and anti-glare coating ensure crisp, clear visuals even in bright indoor lighting. IP66-rated front panel protection, paired with wet touch tracking technology and EN 61000-6-2/6-4 industrial environment EMC compliance, enables long-term, reliable operation in harsh environments, making it an outstanding intelligent management solution for automated factories.Rugged Embedded Computers Zone: Built for Complex and Demanding Industrial EnvironmentsThe Rugged Computing — DIAMOND product line features six product series, all carrying UL safety certification to meet stringent international safety requirements. They support a wide temperature range and a wide voltage input, with industrial-grade protection for high levels of ruggedness and reliability, ensuring worry-free deployment. Customers can select the right model based on desired performance, form factor, power consumption, and industry-specific certification requirements. The newest model, launched in 2026 and the highlight of the show, is the DX-1300 Series, a high-performance compact embedded computer. It supports the latest Intel Arrow Lake-S Core Ultra 200S processor and up to 6400 MHz DDR5 ECC memory, providing ample computing power for Edge AI applications and high-end image processing. Extensive I/O options and diverse expansion capabilities cover all application needs. The compact chassis, combined with a high-end processor and versatile expandability, makes it particularly well-suited for applications that demand high performance in space-constrained installations.GPU Embedded Computers Zone: Full Coverage for Every Edge AI Workload, from Light to HeavyThe GPU Computing — GOLD product line features three series designed to meet different tiers of AI application needs. The GJ Series is purpose-built for entry-level, Light AI applications, featuring NVIDIA Jetson SoM GPUs that deliver efficient AI acceleration with minimal power consumption. For mid-range AI workloads, the GM Series, which supports MXM GPU modules, is the recommended choice. On display is the award-winning GM-1100 Series, recipient of the 2025 Red Dot Design Award and the 2026 Taiwan Excellence Award, delivering outstanding computing performance in a compact chassis, combined with industrial-grade ruggedness and environmental resilience for autonomous mobility equipment, automotive, and image recognition applications. The GP Series excels at Heavy AI workloads, supporting Intel Core CPUs and up to two 300W full-length GPU cards. Three patented designs for expansion, thermal management, and vibration-resistant locking address key user pain points and help customers tackle the most complex and demanding Edge AI applications.

Nuvoton Technology Corporation announced that it has agreed to collaborate with Qualcomm Technologies, Inc. on system-on-chip (SoC) products primarily targeting tethered XR Glass applications. Through this collaboration, Nuvoton will combine its tethered XR Glass–optimized SoCs with Snapdragon XR platforms, and its software ecosystem to accelerate the development and deployment of tethered XR Glass applications and to support industry expansion.XR glasses are increasingly being adopted, particularly in industrial and enterprise fields, for use cases such as operational support, remote assistance, and information visualization. At the same time, achieving compact and lightweight designs, long battery life, and comfortable user experience remains a key challenge for broader market adoption.Nuvoton has been providing SoCs that balance the processing performance and low power consumption required for tethered XR Glass applications. Through this collaboration with Qualcomm Technologies, Nuvoton aims to create an environment that enables more device manufacturers and developers to more easily enter the tethered XR Glass industry.Qualcomm Technologies'newest XR platform, Snapdragon Reality Elite, combined with Nuvoton's SoC, enables a practical and scalable tethered optical see-through (OST) architecture that balances performance, power efficiency, and system flexibility. This configuration allows computing intensive XR workloads to remain on the tethered device while delivering high quality, low latency display and sensor data to lightweight glasses. Together, Snapdragon Reality Elite and Nuvoton’s SoC form a strong foundation for OEMs to accelerate time-to-market for tethered OST designs while maintaining a clear path for future system evolution.Key Collaboration AreasEvaluation of optimized tethered XR Glass configurations by combining Nuvoton's SoCs with Qualcomm Technologies' XR platforms and related technologies.Nuvoton positions its SoC as a key enabler of AI-driven tethered XR platforms. The SoC provides a scalable architecture that decouples application processors from display and sensor subsystems, including cameras and microphones.By abstracting complex interfaces, it helps device manufacturers accelerate time-to-market, reuse designs across product generations, and optimize system performance and power efficiency, particularly for smart glasses.In addition to tethered XR Glass applications, the SoC supports emerging use cases such as agentic AI experiences by simplifying system integration and enabling flexible system architectures. Within the XR ecosystem, Nuvoton works with Qualcomm Technologies, OEMs, and partners at a platform level, supporting the evolution of XR as a next-generation AI computing platform.Value Proposition and BenefitsThrough this collaboration, the adoption of Nuvoton's XR Glass SoCs is expected to deliver the following benefits:1.A comfortable user experience enabled by high-performance, low-power designs optimized for tethered XR Glass use cases. 2.Superior connectivity and access to a rich software ecosystem through compatibility with Qualcomm Technologies platforms. 3.Reduced development effort and shortened time-to-market. 4.Support for a wide range of applications, from industrial to enterprise use cases."This collaboration with Qualcomm Technologies represents an important step in expanding the applicability of our SoC solutions for tethered XR Glass and emerging AI-driven XR platforms." said Ken Su, Vice President of Cloud Security Business Group at Nuvoton Technology Corporation."By working together within the ecosystem, we aim to support our customers in bringing innovative smart glasses and next-generation devices to market more efficiently.""By working with Nuvoton to pair Snapdragon XR platforms with their optimized SoC, we aim to support lightweight optical see-through designs that balance performance, power efficiency and system flexibility. This collaboration builds toward the next phase of the XR ecosystem." Sahil Bansal, Senior Director, Product Management at Qualcomm Technologies remarked.For more information.

SK hynix Inc. (or "the company", www.skhynix.com) announced today that it has shipped samples of HBM4E, a next-generation DRAM for AI, to major customers."The company was able to deliver samples of the 12-stack HBM4E on schedule thanks to its advanced HBM development and production expertise for HBM," said SK hynix, adding that "We will work closely with partners for mass production in a timely manner." The 12-layer HBM4E shows improvements in both performance and power efficiency. The product features a maximum data processing speed of 16Gbps per pin and power efficiency that is up more than 20 percent from previous models. These enhancements improve data processing capabilities for AI training and inference.The HBM4E reduces data transfer latency through its latest interface and design optimization while maintaining stable operation in high-bandwidth environments. This enables customers to increase efficiency in processing data for AI datacenters and large-scale computing systems. SK hynix utilizes Advanced MR-MUF technology for HBM4E products to achieve a 48GB capacity in a 12-layer stack while ensuring structural stability. In particular, the company has also improved heat resistance by 17 percent, compared to the preceding HBM4, enabling stable operation of memory chips in high-performance computing environments.SK hynix has successfully supplied optimized memory solutions to customers based on its expertise in the mass production and supply of HBM3, HBM3E, and HBM4. Leveraging its market-proven product reliability and supply capabilities, the company will support the development of next-generation infrastructure while helping address AI system bottlenecks. "SK hynix has laid the foundation to strengthen its AI leadership with HBM4E based on its market-leading technological capabilities and manufacturing expertise," said Ahn Hyun, President and Chief Development Officer, adding, "Through close collaboration with our partners, we will deliver the value needed in the market while reinforcing our technology leadership as a full-stack AI memory creator."Credit: SK hynix

Advanced Chip and Circuit Materials today announces the commercial availability of Celeritas HM50 and Celeritas HM001, which eliminate the root causes of warpage, package bow, solder fatigue, and high-frequency signal loss in large-format AI accelerators and advanced chip packaging architectures. Celeritas HM50 is a negative CTE (-8 PPM/°C) material and Celeritas HM001 is a near-zero CTE material. Used together in a single stackup, they bring board CTE below 10 ppm/°C while simultaneously delivering Tier 9 electrical performance.Every hyperscaler building AI infrastructure is confronting the same pair of converging constraints. As AI accelerators scale beyond reticle limits, thermomechanical mismatch between silicon (2–4 ppm/°C) and standard PCB materials (~18 ppm/°C) produces catastrophic reflow warpage, package bow, and solder joint fatigue. Simultaneously, the explosive growth in data rates, driven by HBM, UCIe, and chip-to-chip interconnect operating at 100+ Gbps, is pushing signal integrity requirements beyond what standard PCB dielectrics can support. The industry has been searching for two separate solutions to two separate problems. ACCM has built one material family that addresses both.The industry's leading proposed fix of solid glass substrates remains positive in CTE and does not address the electrical side at all. ACCM today announced Celeritas HM50 and Celeritas HM001, which together solve both problems simultaneously. Programs can start today.Celeritas HM50 FEA – Standard PCB at 18 ppm/°C (left) vs. PCB with HM50 at 10 ppm/°C (right). FR4 PCB baseline fails JEDEC qualification, while the PCB with HM50 shows >100× improvement. Warpage and Package Bow are reduced by 64%, and 81%, respectively. Combined HM50+HM001 stackups achieve even lower effective CTE.Keshav Amla, COO of Advanced Chip & Circuit Materials, said,"Rather than incrementally tuning stackups, we are applying a breakthrough materials innovation to remove a fundamental limitation that has constrained system scaling. HM50, with its negative CTE of -8 PPM/°C, drives the effective CTE of the board down. Even with heavy copper designs, you could tune a board down to 12, 10, 8 PPM/°C or lower. And where next-generation data rates demand extreme loss performance, HM001 replaces those layers with a Tier 9 loss material that has a near-zero CTE. Together, they give designers headroom they simply have not had before."The HM Class of materials matches each layer type in an AI accelerator stackup with a material purpose-built for it: HM50 for the power planes, HM001 for the signal layers. As AI accelerators grow in scale, the industry has long struggled with two separate problems — boards warping under thermal stress, and signal loss at extreme data rates. ACCM's Celeritas material family tackles both within a single solution. Celeritas HM50 counteracts the thermal expansion mismatch that causes warpage and solder joint failures, enabling designs that previously failed industry qualification to now meet it with significant margin. Celeritas HM001 addresses the signal integrity side, supporting the data rate demands of next-generation AI interconnects while also contributing to thermal stability. Together, the two materials give chip and board designers headroom that standard PCB materials have not been able to provide. For more details, please visit here.Credit: Advanced Chip and Circuit Materials