Tracking devices play a critical role in the Internet of Things (IoT) by enabling real-time monitoring and data collection across various industries. As IoT applications expand, the demand for energy-efficient solutions becomes increasingly important. This article explores the rise of battery-efficient tracking devices and their transformative impact on IoT ecosystems.Advancements in Tracking DevicesTracking devices first emerged with the advent of GPS technology, revolutionizing location-based services. Early applications were limited, but as the Internet of Things (IoT) grew, tracking devices expanded into sectors like logistics, agriculture, and personal safety, offering real-time monitoring.However, traditional devices faced challenges like short battery life and high maintenance costs. To address these issues, advancements in energy-efficient technologies, such as low-power wireless communication and optimized sensors, led to the development of battery-efficient models. These innovations have paved the way for longer-lasting and cost-effective tracking devices that are essential for the expanding IoT landscape.How Battery-Efficient Technology WorksBattery-efficient tracking devices use low-power wireless technologies like LoRaWAN, NB-IoT, and Bluetooth Low Energy (BLE) to enable long-range communication with minimal power consumption. Sleep modes, energy harvesting techniques (such as solar or kinetic energy), and ultra-low-power processors help extend battery life.Advances in sensor technology allow these devices to collect accurate data while consuming less energy. For example, a GPS trailer tracking uses these technologies to provide long-lasting performance to ensure reliable location tracking without frequent recharging. This makes it ideal for industries like logistics that require continuous monitoring with energy-efficient solutions.Impact on IndustriesBattery-efficient tracking devices have a significant impact across various industries:In logistics and fleet management, they reduce maintenance costs and extend tracking durations for vehicles.In agriculture, smart farming solutions leverage these devices to monitor livestock and machinery with minimal energy consumption.Personal safety benefits from wearable devices that track health and location, requiring less frequent charging.Additionally, the environmental impact of these devices is profound, as longer-lasting batteries help reduce waste and the environmental concerns associated with battery disposal.Benefits Over Traditional Tracking DevicesBattery-efficient tracking devices offer several key advantages over traditional models. Longer operational life is a major benefit, as these devices can run for months or even years on a single charge, reducing the need for frequent recharging or battery replacements. This is especially valuable in remote or hard-to-reach locations.Additionally, reduced maintenance requirements lower operational disruptions, as there’s less need for battery changes or regular maintenance. Lower operational costs are another advantage, with businesses saving on battery replacements and charging infrastructure.Lastly, increased accessibility and scalability enable businesses to deploy more devices across large-scale operations, such as fleet management or asset tracking, without incurring high costs. These benefits make battery-efficient tracking devices an essential solution for industries looking to optimize their IoT networks and reduce long-term costs while enhancing operational efficiency.EndnoteBattery-efficient tracking devices are playing an important role in making IoT systems more sustainable and practical. As technology continues to improve, these devices will become even more integrated into various industries, helping to streamline operations and reduce energy consumption. The ongoing development in this area promises a future where tracking solutions are efficient and cost-effective. 

 SK hynix Inc. (or "the company", www.skhynix.com) announced today that it has begun mass production of the 192GB SOCAMM2, a next-generation memory module standard based on the 1cnm process (sixth-generation of the 10-nanometer technology) LPDDR5X low-power DRAM.SOCAMM2 is a module that adapts low-power memory – which was previously used mainly in mobile products like smartphones – for server environments. It is designed to be a primary memory solution for next-generation AI servers.SK hynix emphasized that the 1cnm based SOCAMM2 product that is now in mass production delivers more than double the bandwidth with over 75% improved power efficiency compared to conventional RDIMM, providing an optimized solution for high performance AI operations.In particular, the company noted that its SOCAMM2 products are designed for NVIDIA Vera Rubin platform.SK hynix expects the new SOCAMM2 product will fundamentally resolve the memory bottlenecks encountered during the training and inference of large language model (LLM) with hundreds of billions of parameters, thereby playing a pivotal role in dramatically accelerating the processing speed of the overall system.The company stated that with the AI market shifting focus from inference to training, SOCAMM2 is gaining significant attention as a next-generation memory solution capable of operating LLMs with low power consumption. To meet the demands of its global Cloud Service Provider (CSP) customers, SK hynix has not only been providing a supply portfolio, but also stabilized its mass production system early on."By supplying the 192GB SOCAMM2, SK hynix has established a new standard for AI memory performance" Justin Kim, President & Head of AI Infra (CMO, Chief Marketing Officer) at SK hynix said. "We will solidify our position as the most trusted AI memory solution provider, through close collaboration with our global AI customers."SK hynix 192GB SOCAMM2, high-bandwidth and power-efficient. Credit: SK hynix 

Molex, a global electronics leader and connectivity innovator, has completed the acquisition of Smiths Interconnect, a subsidiary of United Kingdom-based Smiths Group plc, marking a major milestone in Molex's vision to enable technology that is transforming the future and improving people's lives. Smiths Interconnect brings a diverse portfolio of complementary products and advantaged capabilities in ruggedized custom connectors, contacts, RF components and optical transceivers for harsh environments. Smiths Interconnect also brings leading semiconductor test capabilities that complement Molex's data communications and data center solutions to support growth in AI, along with a strong medical interconnect capability to support Medtech customers."We are excited to finalize the Smiths Interconnect acquisition, which expands our portfolio and engineering expertise worldwide," said Joe Nelligan, CEO, Molex. "Along with their products and technology, we are excited about the Smiths Interconnect's global team and the strong customer relationships that they have established."As the largest acquisition in Molex's history, the addition of Smiths Interconnect positions Molex to drive innovation across the high-reliability interconnect market - encompassing aerospace and defense, space exploration, industrial, medical and semiconductor test industries. The combination of Smiths Interconnect's capabilities with Molex's global footprint, engineering strength, culture and financial strength enables Molex to deliver increased value to customers around the world."This acquisition reinforces Molex's position as a leader across every sector where high reliability is critical," said Michael Cole, SVP and president, Aerospace and Defense Solutions Division, Molex. "By unifying these high-reliability solutions under the Molex banner, we provide a borderless platform that enables engineers to leverage ruggedized, precision-engineered connectivity."The Smiths Interconnect acquisition expands Molex's global footprint to over 90 plants across 22 countries, and its global headcount to over 55,000 employees. The integration creates strong design and technical alignment to enhance innovation and solution capabilities to provide more value to customers. This news builds upon the November 2024 acquisition of AirBorn by Molex, a foundational milestone in serving customers in the aerospace and defense market."The integration of Smiths Interconnect into the Molex engineering ecosystem is a significant win for the global design engineering community," said Lew LaFornara, SVP, Global Interconnect Business Strategy, TTI. "As a prominent Smiths Interconnect distributor, we are excited to continue offering our customers access to a robust line of mission-critical solutions - spanning aerospace and defense, medical, industrial and semiconductor test - all backed by Molex's world-class logistics and technical support."

The rapid proliferation of generative AI, coupled with increasingly sophisticated fraud and disinformation tactics, has made technology-driven risk mitigation and information verification a global priority. Organized by DIGITIMES and co-organized by Taiwan's National Development Council (NDC), the hackathon brought together developers to address these challenges through an "AI against AI" approach.Powered by Amazon Web Services (AWS), the event also featured BitoPro, a cryptocurrency exchange under BitoGroup, and Gogolook, a TrustTech company, as challenge providers. Participants leveraged advanced AI tools to tackle critical societal issues, including fraud prevention for the elderly and virtual currency transaction security, translating innovative concepts into practical and scalable solutions.Fraud Prevention Elevated to National Security Priority: Government Strengthens Cross-Agency CollaborationDuring the awards ceremony, Tsai Yen-ming, Deputy Commissioner of the Criminal Investigation Bureau (CIB) under Taiwan's National Police Agency (NPA), emphasized that fraud has evolved from isolated criminal incidents into a complex and systemic threat-undermining social trust, disrupting financial order, and even posing risks to national security.To address these challenges, the CIB has established a Technology Crime Prevention Center to strengthen technology-driven investigative capabilities and talent development, alongside a Fraud Prevention Center that integrates cross-ministerial resources coordinated by the Executive Yuan to enhance overall anti-fraud capacity. Tsai noted that the competition's focus on fraud prevention and virtual currency transaction security has already demonstrated strong potential for real-world implementation. He expressed optimism about deepening collaboration with private-sector teams to accelerate the adoption of AI-driven anti-fraud technologies within the public sector.Empowering the Next Generation of Tech TalentThe hackathon attracted a diverse group of young developers from universities across Taiwan. During the event, Colley Huang, Chairman of DIGITIMES and IC Radio, encouraged participants to identify their roles and create value amid a rapidly evolving technological landscape. He noted that today's generation faces both challenges and opportunities shaped by the convergence of AI, semiconductors, and geopolitics. Early engagement with emerging technologies and the cultivation of differentiated capabilities, he said, will be essential for building long-term competitiveness.AWS Provides Cloud Infrastructure and AI Tools to Enhance Solution FeasibilityTo enhance the feasibility of the proposed solutions, AWS provided a comprehensive cloud development environment and technical support. Prior to the competition, participants attended workshops to familiarize themselves with AWS's advanced cloud services and AI capabilities. Teams also utilized Kiro, a purpose-built IDE with advanced AI agents, enabling them to rapidly build robust minimum viable products (MVPs) within the 27-hour development period.Robert Wang, Managing Director of AWS Taiwan, stated that the hackathon successfully brought together government, enterprises, and developers specifically focused on fraud prevention - a highly critical social issue - demonstrating the power of cross-sector collaboration in leveraging technology to solve complex challenges. He noted that fraud tactics continue to evolve by exploiting both technological advancements and human vulnerabilities, making it essential to deploy "equivalent countermeasures" powered by AI and data analytics. Wang further highlighted that Taiwan's extensive experience in fraud prevention holds significant export potential, and that the competition's outcomes not only address domestic challenges but also underscore the global commercial viability of Taiwan-developed anti-fraud solutions.Gogolook: Building a Cross-Sector Anti-Fraud EcosystemIn the "Fraud Identification and Prevention" category, with Gogolook serving as the challenge provider, many teams extensively leveraged Amazon Bedrock, a fully managed AWS service that provides access to high-performing foundation models. By integrating multiple foundation models, participants elevated fraud prevention from simple alert notifications to sophisticated, logic-driven explanations that help users understand why content is fraudulent. Team (1) demonstrated the practical application of AI in enhancing real-time fraud detection and decision-making, ultimately securing the top award.Reiny Song, Co-founder and CTO of Gogolook, remarked that fraud has evolved into a systemic issue spanning digital ecosystems, financial systems, and societal trust. While AI has amplified emerging risks, it also represents a critical tool in combating fraud. He emphasized that effective fraud prevention requires more than a single product; it depends on building a cross-sector ecosystem that integrates technology, real-world scenarios, and talent to drive scalable implementation. The competition showcased not only innovative concepts but also practical and viable technical solutions, advancing fraud prevention from theory to execution.BitoGroup: Strengthening Security in the Virtual Currency EcosystemIn the "Virtual Currency Transaction Security" category, Team "Bibi Lab" secured the top award. Titan Cheng, Founder and CEO of BitoGroup, the challenge provider for this category, noted that fraud and cybersecurity threats will become increasingly complex in the AI era, particularly as virtual currency-related crimes continue to evolve. Addressing these challenges, he said, will require sustained and high-intensity efforts rather than short-term initiatives.Cheng emphasized the importance of strengthening data governance and risk detection capabilities through the exchange of technical frameworks and practical experience, enabling both enterprises and society to collectively elevate their defense capabilities. He added that the competition not only served as a platform to showcase and validate technological innovations but also helped drive the broader adoption and advancement of anti-fraud solutions across the industry.Advancing Taiwan's AI and Bilingual Talent DevelopmentThe hackathon aligns with Taiwan's national agenda to promote AI industrialization and enhance digital societal security, while strengthening international collaboration and talent cultivation. The event received support from the National Development Council's subsidy program for promoting bilingual talent development and fostering an English-friendly environment.By convening nearly 100 AI professionals from across Taiwan, the competition translated the government's "fraud awareness for all" initiative into scalable and practical technological solutions, enhancing public awareness and resilience against emerging fraud tactics.The "Agent for Truth: Disinformation Defense Hackathon" served not only as a technological competition but also as a public-wide collaborative effort to build a digital defense network through public-private partnership and cross-generational co-creation. By bringing together cross-disciplinary AI talent to develop innovative solutions that genuinely protect citizens and strengthen societal resilience, the event demonstrated the potential for deeper cross-sector collaboration to build a more robust trustworthy information ecosystem and create a safer, more digitally resilient information environment for Taiwan.Government and industry leaders gather at the hackathon awards ceremony. Credit: DIGITIMES

Amidst the reality of global oil prices continuously skyrocketing due to war, the global market is facing the formidable challenges posed by the turbulence impacting the global economy in 2026. Major Taiwanese electronics manufacturers and supply chains are pivoting to review comprehensive AI strategies. They are adopting a particularly aggressive stance in their strategic deployment of cloud server systems and AI infrastructure, aiming to achieve an operational breakthrough where the market outlook remains far from optimistic.ASUS unveils its complete Professional Services for Sovereign AI. Since the first project in 2018, which resulted in Taiwan's first AI supercomputer, Taiwania 2, ASUS has been actively developing sovereign AI infrastructure and providing customized AI solutions to clients. Delivering trusted AI with total flexibility, from rack-scale AI Factories, desktop AI supercomputing, Edge AI to Enterprise AI solutions deployment, the company is redefining Sovereign AI Solutions with reinforces the company's position as a global leader in AI-driven digital transformation.On April 1, 2026, ASUS joins the DIGITIMES online seminar titled "Leading the Way in Sovereign AI: A Look at Next-Generation AI Computing and Storage Architecture from Supercomputing Performance". This event starts with the construction and practical application of next-generation supercomputers at the National Center for High-Performance Computing (NCHC) of the National Applied Research Laboratories. This collaborative project shows ASUS Powering NCHC to building Taiwan's AI Supercomputer and the next generation of High Performance Computing (HPC) systems.As AI and high HPC workloads push compute density and power consumption beyond the capabilities of traditional air cooling, the flagship ASUS offerings are the ASUS AI POD systems redefining the new architecture of rack-scale powerhouse, storage architecture, and liquid cooling designed for massive AI workloads. This momentum will continue to maximize computing power density through the NVIDIA HGXTM and Blackwell platforms and provide insights into solving heat dissipation bottlenecks with advanced liquid cooling technology to meet the performance and energy efficiency requirements of high-computing environments.Practical experience in building a leading infrastructure in Taiwan High Performance ComputingThe first session features a keynote lecture by Nobel Hsia, ASUS Deputy Manager of Product Planning, titled "Leading the Sovereign AI Wave: From the National Center for High-performance Computing (NCHC) projects to ASUS's Forward-Looking Infrastructure and Storage Solutions." ASUS's Sovereign AI targets the business opportunities that countries are currently striving to play the vital role of innovation in securing data sovereignty and security with in-house data processing and computing capabilities, and promotes the transformation of scientific research in the industry. ASUS has partnered with NVIDIA to build a massive AI supercomputer system with an "ALL IN on AI" strategy. Through the maximum computing power density provided by the NVIDIA HGXTM and Blackwell platforms, ASUS has deployed complete portfolio ranging from rack-scale AI factories to edge and enterprise deployment. With its proven track record of successfully supplying AI infrastructure at the B300 and GB300 levels in the market, ASUS has become a trusted partner, providing complete end-to-end solutions.It is worth mentioning that ASUS has ten years of experience collaborating with the National Center for High-performance Computing on supercomputer projects. ASUS's sovereign AI expertise is proven by several successful national-level AI project deployments, including flagship supercomputing initiatives such as Taiwania 2 and Forerunner 1, cementing its leadership in high-performance computing and AI systems. Through ASUS Professional Services, the collaboration covers all aspects from design, deployment to operation.The latest Nano4 (Crystal 26) NVIDIA HGXTM H200 cluster AI server system was to build a next-generation AI supercomputer capable of handling complex large language models (LLMs), deep learning, and advanced HPC workloads. In this new project ASUS has built Taiwan's first AI supercomputer project based on NVIDIA GB200 NVL72 system architecture with direct-cooling technology. The engineering team has played an important role in the delivery of the servers, the planning of large-scale computing architecture, deployment and optimization, which has become the foundation for ASUS to expand its overseas AI deployment and demonstrate ASUS's ability to build computing power from national to international levels.Opening Agentic AI Frontier: ASUS Supports NVIDIA Vera Rubin Platform and InfrastructureIn response to the advent of the new-generation NVIDIA Vera Rubin platform and its accompanying infrastructure architecture, ASUS has developed the next-generation ASUS AI POD highlighting a proficiency in liquid-cooled AI solution. Spanning the new design from rack-scale AI factories, data center servers to desktop workstations and edge AI devices, this solution delivers end-to-end AI computing power and infrastructure, specifically targeting trillion-parameter models and million-token contexts, maximizing efficiency across power, memory, and compute.Hsia began by introducing the ASUS flagship XA VR721-E3 architecture. This system supports the NVIDIA Vera Rubin NVL72 platforms; it is purpose-built for large-scale AI model inference and training to deliver massive AI performance for large-scale AI factories. while also accommodating the specific workloads required for agentic AI. Designed as a 100% liquid-cooled, rack-scale system, it features a Thermal Design Power (TDP) reaching up to 227 kW, which is a capability that fully satisfies the demand for the immense performance required to compute AI models with trillions of parameters.Furthermore, addressing rigorous enterprise-grade data-center demands, ASUS has simultaneously launched the XA NR series. These product series support the NVIDIA HGXTM Rubin NVL8 architecture, featuring eight Rubin GPUs interconnected via the sixth-generation NVLink, which each GPU could deliver a maximum bandwidth of 800 GB/s. To facilitate a seamless and cost-effective transition to liquid cooling, ASUS offers two distinct solutions: the XA NR1I-E12L, an innovative hybrid-cooled option; and the XA NR1I-E12LR, a 100% liquid-cooled system.To support these powerful systems and democratize AI development, ASUS also has established a robust data ecosystem by partnering with NVIDIA-Certified storage providers. The storage solutions offers the technologies such as JBOD, DPDK, and Object Storage, thereby delivering scalable, resilient solutions for memory-intensive AI applications to support robust capabilities for the integrated storage and operational management.On the software front, ASUS provides a suite of one-stop platforms. Leveraging the ASUS Infrastructure Deployment Center (AIDC), ASUS automated the setup process including ACC (ASUS Control Center) and BMC, accelerating time-to-market for critical research resources. To address the full spectrum of requirements for system construction, deployment, and operations, ASUS provides expert consultation, a broad portfolio of tailor-made AI solution while simultaneously catering to the comprehensive lifecycle management needs of the entire data center.WEKA data storage platform is redefining AI storage economicsTo meet the elastic data storage requirements of AI workloads across diverse usage scenarios, ASUS's AI POD system incorporates a comprehensive storage solution featuring a high-speed, All-Flash NVMe SSD storage architecture. ASUS has collaborated with its ecosystem partners to develop a next-generation unified storage system with high-reliability storage servers and professional validation of network. WEKA, the AI storage company, is one of partners providing high-performance, software-defined storage for GPU-accelerated AI and HPC environments, combining low latency with unified data management.In the second presentation of the session is delivered by Ray Wu, Senior Consultant for the Asia-Pacific region at WEKA. His speech titled the theme "Ultimate Data Empowerment: How WEKA Helps NCHC Build an AI-Accelerated Computing Storage Architecture". To address NCHC's requirements for extreme performance and energy efficiency, Wu highlighted the unified data management solutions to provide rapid scalability, which emphasizing flexibility and intelligent adaptive capabilities, and effectively addressing the demands of a wide array of usage scenarios.Addressing the requirements of diverse application environments such as Kubernetes and Slurm in NCHC Nano4, WEKA serves as the high-performance storage foundation through its efficient storage platform. Based on the NVIDIA AI Data Platform reference architecture, the solution employs an end-to-end system to accelerate data processing performance for HPC applications. This assists the NCHC in optimizing the deployment of high-efficiency AI computing, dramatically reducing the time required for AI application deployment and development from months to mere minutes.This ASUS WEKA storage solutions demonstrates the NCHC's technical excellence in leveraging its ecosystem and integrated services to rapidly deliver the low-latency, scalable and high throughput required capabilities and swiftly support the next generation of agent-based AI applications. WEKA solutions empower enterprises to transition from experimentation to full-scale operations, rendering AI applications economically viable and maximizing performance across a wide spectrum of fields, ranging from next-generation AI agent systems to AI-enabled healthcare applications.In this event, ASUS showcases fully liquid-cooled AI infrastructure with the critical thermal management required for the next-generation NVIDIA Vera Rubin NVL72 system platform. By efficiently dissipating heat from high-performance CPUs, GPUs, and accelerator-dense racks, ASUS significantly reduces energy consumption while supporting unprecedented rack density to enable enterprises and cloud service providers to build high-performance, energy-efficient large-scale AI clusters with unmatched efficiency and dramatically reduced PUE and TCO.【White paper download】Empowering Scalable AI Reasoning with ASUS AI POD featuring【Webinar on-demand】Bridging NCHC Success to NVIDIA Vera Rubin Architecture 

Cypress Technology has been dedicated to the research, development, and manufacturing of professional audiovisual solutions for 36 years, with all R&D and production facilities based in Taiwan. In recent years, the company has actively engaged with various application environments, gaining deep insights through close collaboration with users and on-site evaluations. This approach has enabled the continuous development of integrated solutions that meet the specific needs of diverse scenarios.With the rapid evolution of audiovisual technologies, the HyLinX AV over IP solution which leverages standard network packet transmission has emerged as a key trend in Pro AV industry due to its flexible deployment, centralized management, and high scalability. Cypress Technology has successfully implemented this technology in multiple judicial institutions both in Taiwan and abroad, helping transform courtrooms into efficient and intelligent digital environments.During court proceedings, the ability to present evidence and critical information to judges and juries in a real-time, clear, and transparent manner is essential to ensuring a smooth trial. As a result, comprehensive AV equipment configuration is crucial. Modern courtrooms are typically equipped with high-resolution displays and document cameras to present various types of evidence and documents in real time. Through the professional AV integration solutions of Cypress Technology to enable rapid switching and synchronized display of multiple signal sources, ensuring clear information delivery and intuitive operation.In addition, Cypress Technology's judicial AV integration solution can transmit video and audio signals from identification rooms on different floors to the courtroom in real time. This supports remote interrogation applications, allowing witnesses or involved parties to complete appearance procedures without being physically present. Through high-quality video transmission and synchronized recording functions, the entire trial process can be fully preserved, enhancing the convenience and reliability of judicial operations.Furthermore, the system displays evidence and important information simultaneously on high-resolution equipment, allowing judges and juries to grasp key content in real time. By integrating remote video participation with evidence retention, the solution not only ensures information integrity but also enhances trial efficiency and overall workflow.Beyond judicial applications, Cypress Technology's professional AV solutions are widely utilized in high-performance corporate meeting rooms, command and control centers, smart manufacturing situation rooms, medical imaging transmission, as well as education and public display environments. Moving forward, Cypress Technology will continue to invest in the research and development of high-efficiency audiovisual technologies while integrating sustainability into its product design and manufacturing processes. Cypress Technology remains committed to providing stable and reliable solutions that enable smarter and more efficient audiovisual workflows across diverse applications.Evidence Presentation in Digital Courtroom. Credit: Cypress Technology

STMicroelectronics (NYSE: STM), a global semiconductor leader serving customers across the spectrum of electronics applications, today announced the expansion of its 800 VDC power conversion portfolio with two new advanced architectures: 800 VDC to 12V and 800 VDC to 6V. Developed according to the NVIDIA 800 VDC reference design, these new power conversion stages complement the previously introduced 800 VDC to 50V solution. The rapidly emerging 800 VDC data center architecture enables higher energy efficiency, reduces power losses, and supports more scalable, high compute density, infrastructure for hyperscalers and AI compute."As AI infrastructure compute scale continues to expand fast, it requires higher voltage distribution and greater density, which can only be achieved with system-level innovation for each of the different AI server form factors," said Marco Cassis, President, Analog, Power & Discrete, MEMS and Sensors Group Head of STMicroelectronics' Strategy, System Research and Applications, Innovation Office at STMicroelectronics. "With these new converters for 800 VDC power distribution, ST brings a complete set of solutions to support the deployment of gigawatt-scale compute infrastructure with more efficient, scalable, and sustainable power architectures."A complete 800 VDC ecosystem for the different AI server form factorsThe expansion to 12V and 6V output stages reflects the industry move toward different server architectures requiring different power delivery topologies depending on GPU generation, server height, form factor, and thermal envelope for large-scale training clusters, inference farms, and high-density AI infrastructures. The 50V, 12V, and 6V intermediate DC buses will all coexist in AI data centers depending on rack density, GPU configuration, and cooling strategy.The new 800 VDC to 12V converter enables high-efficiency distribution from rack-level power shelves directly to the voltage domains that feed advanced AI accelerators.The new 800 VDC to 6V path allows OEMs to reduce the number of conversion stages and move the 6V bus closer to the GPU. This reduces copper usage, minimizes resistive losses, and improves transient performance, a critical differentiator for large-scale training clusters.Back in October 2025, STMicroelectronics introduced a fully integrated prototype power？delivery system showcasing a compact GaN？based LLC converter operating directly from 800 V at 1 MHz with over 98% efficiency and exceptional power density in a smartphone？sized footprint exceeding 2,600 W/in³ at 50 V. The three solutions combine ST technologies across power semiconductors (silicon, SiC, GaN), analog and mixed-signal, and microcontrollers. Technical highlights of the new 12V and 6V architecturesDirect 800 VDC to 12V high-efficiency conversion:Eliminates the traditional 54V intermediate stage, reducing conversion steps and system-level losses. Enables higher rack-level efficiency, lower copper usage, and simplified integration for future GPU generations. Includes newly developed high-density power delivery board (PDB) achieving efficiency targets exceeding the sum of previous two-stage conversion paths.800 VDC to 6V architecture for GPU-nearing conversion:Is designed for system builders who require power stages closer to the GPU, minimizing IR drop and improving response under fast load transients.Completes the topology portfolio for servers with ultra dense GPU configurations.Additional technical information at the site.

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.

Karsan, the world's technology-oriented mobility brand in next-generation public transportation, increased its turnover by 11% to 330 million Euros in 2025. Karsan, which generated 220 million Euros of turnover from electric vehicle sales, saw a 43% increase in its exports to 197 million Euros. Aiming to maintain its growth in Europe, Karsan plans to begin driverless autonomous public transportation operations in 2026. Stating that Karsan e-JEST and e-ATAK models preserve leadership in their segments in the European public transportation market, Karsan CEO Okan Bas said, "Turkiye is Europe's production base in bus & midibus manufacturing. Karsan conducted 80% of the electric minibus and bus exports from Turkiye to Europe between 2019 and 2025."Emphasizing that they aim to maintain their growth in Europe by boosting their EV sales in 2026, ranking among the top 5 players in Europe in the next five years, Okan Bas mentioned, "Of course, we are aware that our presence in the existing markets is not sufficient to achieve this target.Accordingly, we will expand into new markets. We will focus on some northern countries in 2026. We do not operate in the Netherlands, Sweden, Norway, or Germany. In these markets, we will first initiate our organizational structuring. We will also increase our strength in the Spanish and Polish markets we entered last year. We aim to expand our electric vehicle fleet in Europe to exceed 2,800 units, representing over 30% growth. Autonomous mobility is among our strategic focal points. We will concentrate on Europe with the 8-meter Autonomous e-ATAK, and on America with the 6-meter Autonomous e-JEST. In the autonomous mobility field, we have started our efforts to remove the safety driver from the vehicle. We aim to initiate fully driverless operations in Stavanger by Q3 2026. Towards the end of 2026, we will add another electric vehicle to the product range. We constantly strengthen our presence by expanding our product range and accelerating our development in technology."Playing a leading role in transforming public transport globally with the vision of "One Step Ahead in the Future of Mobility", Karsan has left behind another successful year. Karsan, which has leveraged strong momentum in public transport with electric and autonomous vehicles, continues to strengthen its global presence through a vehicle park in 27 countries. The brand, which has made a significant contribution to the transformation of public transportation with over 2,100 electric vehicles, has set a higher target for 2026.Karsan e-JEST and e-ATAK maintain leadership in Europe!Noting that they completed the year 2025 with a turnover of 330 million Euros, Karsan CEO Okan Bas said, "We increased our total turnover by 11% compared to 296 million Euros in 2024. Electric vehicles account for 220 million Euros of this figure. Thus, we enhanced our electric vehicle turnover by 59 million Euros. Our electric vehicle turnover, which delivered a significant growth of 37%, also accounted for 67% of our total turnover. Last year, we increased our export turnover by 43 % to 197 million Euros. We boosted our EBITDA from 32 million Euros in 2024 to 54 million last year."Mentioning that they, as Karsan, focus on three technologies: electric, autonomous, and hydrogen, Okan Bas continued: "Since 2021, we have had presence in the market with our 6-8-10-12 and 18-meter full-electric product range. In the European urban electric public transport market, which grew by 39% last year, we could advance one position. In 2025, we increased our market share in Europe by 0.5 points to 5%. Thus, we rose from 8th to 7th in the ranking. During this period, e-JEST became the leader in the electric public transportation minibus market with 30% share, while e-ATAK secured a 25% share, leading the electric midibus class. Karsan e-JEST, which was the leader of its segment for the first time in 2020, has not ceded this title to any of its competitors for 6 years. In 2025, our 10-meter e-ATA model became the leader in the 10-meter electric bus market in Europe by gaining a share of 26%."Goal: Increasing the share of exports in turnover to more than 70%!Underlining that they have ranked among the top four in nine countries across Europe, Okan Bas said, "In this regard, we are the market leader with 34% in Romania and 50% in Croatia. Karsan is the second brand with 12.5% in Italy and 19.5% in Bulgaria. In addition, we are among the third brands in France, Portugal, and Greece, and the top four in Poland and Switzerland." Stating that their brand is the pioneer and leader of the Turkish electric public transportation market, Karsan CEO Okan Bas said: "In 2025, 69 electric public transport vehicles were sold in Turkiye. We, as Karsan, sold all of these. A total of 139 Karsan electric vehicles are deployed in Turkiye. Turkiye is also Europe's production base in bus and midibus manufacturing. Our country has produced one out of every two buses in Europe. Karsan conducted 80% of electric minibus and bus exports from Turkiye to Europe between 2019 and 2025. In 2025, we increased our electric vehicle exports by 15%, reaching 555 units. As of today, the part of our exports in our total turnover is nearly 60%. We intend to raise this rate to more than 70% in three years."Autonomous e-JEST goes to the World Cup!Asserting that Karsan's global electric vehicle park has reached 2,130 units, Okan Bas stated, "As of today, our electric vehicles continue to carry passengers in 27 countries on three continents. In 2025, we expanded our electric vehicle fleet by 40%. We have achieved a very significant success by winning 40% of the tenders we have participated in. We have also gained notable experience and operational competence in autonomous mobility. We have undertaken 16 projects in 12 countries with Karsan Autonomous e-ATAK, the first Level-4 Autonomous bus carrying ticketed passengers globally. 12 of them have been implemented, and 4 will be initiated soon. As a result of these efforts, we have accumulated 160,000 kilometers of road experience, serving 60,000 passengers autonomously. We observe that the demand for autonomous solutions is increasing. In our first three years, we commissioned seven projects, and in 2025, we executed nine new ones. The most substantial factors that increase interest in autonomous mobility are high operating costs, difficulty in finding drivers, and, of course, accidents. Since we saw potential in this field, we developed and unveiled an autonomous version of our e-JEST model. In Autonomous e-JEST, we brought together the four years of experience from Autonomous e-ATAK and the seven years of public transportation experience of e-JEST. After the launch, we immediately received 10 orders for Autonomous e-JEST from the USA. We will deliver these vehicles in the second half of 2026, and they will be used at the World Cup.”Karsan have reached 65% of our sales target!Unveiling Karsan's goals for 2026, Okan Bas said, "We have 60 years of automotive experience. We have reached a power of 2,500 people. I always say that when these 2,500 employees focus on a common goal, a different kind of power emerges. We set up our system to take this advantage. Our primary goal is to grow our electric sales further and increase their share in our total sales to 80%. We wish to grow in 2026 by maintaining a 5% market share. In the medium term, we want to be among the top five brands in Europe in the next 5 years. We, of course, know that our presence in existing markets is not sufficient to do this, and we will expand into new markets. We will focus on some northern countries in 2026. We do not operate in the Netherlands, Sweden, Norway, or Germany. In these countries, we will first initiate our organizational structuring. These are all the most challenging countries in Europe to operate in. We will enter new markets and be stronger in Spain, for instance, which we penetrated last year. We also need to get stronger in Poland, where we have very serious plans for 2026."Reminding that they have allocated some of this year's production for last year's orders, Okan Bas continued: "Today, we have met 65% of our 2026 target with the confirmed orders. We are currently producing these orders. In the meantime, we continue to collect new ones. Although we had nearly no orders 4-5 years ago, today we are entering the new year with a strong order portfolio thanks to our growing business volume. In 2026, we aim to sell approximately 700 electric vehicles. In Europe, we have created plans to boost our electric sales to more than 2,800 units, coinciding with an increase of over 30%. Autonomous mobility is among our focal points.""We will concentrate on Europe with the 8-meter Autonomous e-ATAK, and on America with the 6-meter Autonomous e-JEST. Our goal is to expand and effectively manage the market. There are two critical points: Removing the safety driver from the vehicle and Type approval. In the first half of 2027, we would like to secure Type approval for autonomous vehicles. Our tests to completely phase out the safety driver have also started. We aim to initiate fully driverless operations in Stavanger by Q3 2026.""Karsan has two fundamental strategies. One of them is that, as Karsan, we want to be a strong brand in the global market within this change. The second is to produce on behalf of global brands. We are transforming from a traditional automotive company to a mobility technology company in this evolving world. This is how we position ourselves clearly. In other words, we believe that no traditional automotive company will survive in the long term. This is how we handle this change. In this transformation, we concentrate on where we focus. In addition, towards the end of this year, we will add another electric vehicle to the product range. So, we are constantly expanding our presence through our product range and technological progress. We think that the hydrogen technology will have a share of around 10% in the total bus market, particularly in public transportation. In this regard, we have a strategic partnership with Toyota. We sold our first 2 vehicles to Italy last year and received 3 new orders. In this field, too, we will grow step by step."

DEEPX, a Seoul-based fabless semiconductor company developing ultra-low-power AI inference chips for physical AI applications, has secured 27 commercial purchase orders across eight countries within seven months of starting mass production of its first-generation AI chip - a pace that industry observers describe as highly unusual for an emerging fabless company at such an early stage of commercialization.The orders span seven major Physical AI application domains, including robotics, smart factories, edge AI servers, industrial AI, surveillance, AI IT services, and smart cities, with deployments now active across Asia, North America, and Europe.Global Expansion at Unusual SpeedThe ramp-up was slow at first: only two orders came in during the first five months after production began. But the company added 25 additional orders in less than three months in 2026, indicating rapidly accelerating commercial adoption.While some semiconductor startups have achieved limited international shipments, DEEPX's simultaneous expansion across multiple countries and application domains at this pace is highly unusual, according to industry observers. The company's chips are now being deployed across eight countries spanning Asia, North America, and Europe.Early Global Supply Chain ExecutionDEEPX established a global distribution network early in its commercialization phase through partnerships with Avnet, DigiKey, and WPG. Industry analysts note that it is rare for an emerging fabless semiconductor company to secure a global supply chain at such an early stage of mass production.In Europe, DEEPX signed a distribution agreement with Avnet Silica - the European arm of Avnet - establishing a regional supply chain ahead of broader market entry. Avnet Silica has already identified more than 30 prospective customers across high-performance embedded segments including smart city infrastructure, autonomous mobile robots (AMR), machine vision, and smart factories. Both companies are actively expanding purchase contracts as European industrial demand for edge AI inference accelerates.Partner Ecosystem: From Silicon Vendors to AI Software LeadersA key driver of DEEPX's commercial momentum is a cross-industry partner ecosystem spanning semiconductor vendors, industrial computing platforms, and open-source AI software leaders - reflecting a deliberate strategy to position its silicon as an embedded AI standard rather than a point solution.Renesas Electronics: Has integrated DEEPX NPUs into more than three types of industrial boards combining Renesas application processors, delivering solutions applicable to smart factory deployments. The collaboration between a tier-one AP semiconductor company and an AI chip startup has been widely noted as an uncommon signal of hardware ecosystem maturity.Sixfab and Raspberry Pi: Have jointly developed and released an AI HAT module built around the DX-M1, alongside a real-time smart traffic analysis solution using the CES 2026 Best of Innovation-winning 'ALPON X5' AI PC that integrates DEEPX products.AAEON (ASUS subsidiary), IEI, WeLink, Endrich, Toradex, and Lanner: Industrial system developers have deployed customized AI hardware solutions for smart cities, automated logistics, and security systems powered by DEEPX chips, extending the company's reach across diverse embedded application segments.Ultralytics & Network Optix: Ultralytics, developer of the widely-adopted YOLO vision AI model family, has demonstrated 'Open-Source Physical AI Alliance' workloads running on the DEEPX NPU with a one-click deployment flow. Network Optix has validated an intelligent Video Management System capable of managing thousands of camera feeds simultaneously using DEEPX inference.New Products: DX-M1M and DX-AIPlayerAlongside its commercial order momentum, DEEPX has launched two mass-production product lines designed to reduce integration friction for industrial developers across its key markets:DX-M1M (M.2 Module): A standard M.2 form factor module embedding the DX-M1 AI chip, designed to drop into existing industrial PCs and edge servers. The format allows customers to add powerful AI inference capability without redesigning hardware, significantly lowering the barrier to deployment.DX-AIPlayer: An edge AI acceleration solution integrating the DEEPX NPU with a high-efficiency CPU board. Positioned as a 'one-stop' platform covering the full workflow from model development to industrial deployment, the product enables developers to implement vision AI applications without complex integration overhead.Large-Scale Global PoC StrategyThe company attributes its rapid adoption to a large-scale pre-production engagement strategy. In the semiconductor industry, it typically takes 9 to 18 months from PoC initiation to full production deployment in customer applications. To shorten this cycle, DEEPX began working with customers well before mass production began.Over more than one year prior to production, the company collaborated with approximately 350 global companies, supporting proof-of-concept testing and system integration. This approach created a large pipeline of customers already familiar with the silicon and ready to transition into commercial orders once production began.Physical AI Moving from Concept to RealityCEO Lokwon Kim pointed to the growing volume of PoC activity as the clearest signal that physical AI is no longer a forward-looking concept. "Many people still view Physical AI as a concept for the future. But when we look at the growing number of PoC projects with global companies today, it becomes clear that Physical AI is already becoming a reality," he said.On the commercial order milestone, Kim was direct. "Securing 27 commercial orders within seven months is one of the early signs that this transition is already happening in the industry." With distribution in place across Asia, North America, and Europe, and a partner ecosystem spanning silicon vendors, industrial platform makers, and open-source AI software leaders, DEEPX appears positioned to expand its commercial footprint across multiple regions simultaneously. If the current order velocity holds, the company's ambition to establish its NPU as the default physical AI inference chip for embedded developers worldwide may be closer to reality than the industry expected.