SoftBank Group (SBG) has denied SpaceX CEO Elon Musk's claim on X that the company defaulted after ordering turbines, while saying its AI infrastructure plans remain on track. However, SBG recorded a JPY597.2 billion (US$3.73 billion) loss in the second quarter of fiscal 2026 (July–September 2026) after reviewing purchases of power-generation equipment for US data centers, though it said the loss was unrelated to Project Stargate Ohio.

OpenAI CEO Sam Altman warned Samsung Electronics executives that AI memory demand could fluctuate in the short term, with the market potentially counting the same demand more than once. He nevertheless expects long-term AI demand to surge, creating shortages of chips and data centers and underscoring the need for greater supply chain investment.

Meta Platforms' longtime infrastructure chief Santosh Janardhan will leave after the end of 1Q27, with former Amazon Web Services (AWS) vice president Dave Brown set to succeed him. The leadership transition comes as Meta accelerates its AI strategy and expands data centers and in-house chip development.

Chinese AI models have taken a dominant position as demand for open-source models grows among US businesses. According to the "2026: The State of Enterprise AI" report released in early November by US venture capital firm Menlo Ventures, Chinese models accounted for 89% of US corporate spending on open-source AI models.

As demand for AI computing expands across industries and applications, chipmakers are seeking opportunities beyond conventional data centers, extending into enterprise deployments and personal devices. At the ninth China International Import Expo (CIIE) in 2026, Intel China showcased three AI computing solutions: green computing pods for large data centers and enterprise campuses, integrated AI inference systems for small and medium-sized enterprises (SMEs), and AI Agent PCs for individual users.

At the opening panel of electronica 2026 in Munich on November 10, industry leaders agreed that the primary bottleneck in artificial intelligence is shifting from computing power to energy delivery. Panelists warned that power semiconductors are quickly becoming the tech sector's next critical supply chain gap as AI data centers scale to hundreds of megawatts, or even gigawatts, straining grids and power delivery architectures never built for such concentrated loads.

A semiconductor chip no thicker than a fraction of a human hair might seem insignificant beside the powerful silicon processors driving the world's AI revolution. Yet at Pragmatic Park in North East England, these tiny, flexible devices represent a different vision of computing: one where intelligence gradually becomes embedded in the objects people touch, wear, and use every day.

The award ceremony for the fourth K. T. Li Award was held in Taipei on November 10, with this year's awards presented by former President Ing-wen Tsai to Kinpo-Compal Group president Rock Hsu and Quanta Computer founder Barry Lam.

Nvidia CEO Jensen's Huang's affair with an alleged Chinese spy has raised questions about national security, corporate governance, and even whether he will keep his job. Beyond his outsized role within the company, the case has also brought attention to his family members' roles within the firm and their influence over a public company that is currently the singular force behind the AI revolution.

TECO Electric & Machinery is expanding into global AI computing infrastructure, announcing on the November 11 that it will partner with AI cloud service provider GMI Cloud and KSDI to build an modular AI data center in Satsumasendai, Kagoshima Prefecture, Japan. TECO will serve as the modular data center contractor, handling design, electromechanical systems integration, and technical guidance, with the first 6.4MW phase scheduled for delivery within nine months.

Demand for AI, high-performance computing (HPC), and semiconductors continued to rise, driving Wah Lee Industrial's 3Q26 profit to record-high levels. Gross profit reached NT$2.06 billion (US$62.7 million), up 28.7% year-on-year, while earnings per share (EPS) reached NT$3.46; cumulative EPS for the first three quarters rose to NT$9.4, also a record for the period.

US silicon carbide (SiC) giant Wolfspeed will re-enter the gallium nitride (GaN) market through a vertically integrated model and expand into the GaN-on-silicon (GaN-on-Si) power device market, according to company CEO Robert Feurle. The announcement, made at a media event during the Electronica 2026 trade fair in Munich, Germany, stated that the company will initially target low-voltage applications below 650V.

Zhen Ding Technology said strong seasonal demand for mobile communications products and robust demand for high-end AI applications drove its third-quarter 2026 revenue to a record high. Rising shipments of high-end products and an optimized product mix also lifted gross margin 5.7 percentage points year-over-year to a record 27.7%.

The FBI has launched an investigation following an alleged extramarital affair involving Nvidia CEO Jensen Huang and Alice Huang, a former executive at Singapore-based Megaspeed who has been accused of having ties to Chinese intelligence. Taiwan's Ministry of the Interior said it has revoked Alice Huang's entry documents over the espionage allegations, barring her from entering Taiwan in the future.

The US midterm election results, which delivered a major Democratic victory, and national security concerns surrounding an affair between an alleged Chinese spy and Nvidia CEO Jensen Huang have raised fears that AI data-center construction in the US could be disrupted. While few in the industry doubt AI's long-term growth, the impact of funding displacement warrants close attention.

Global AI data center construction continues to drive demand for GPUs, memory, and other critical hardware, but tight computing capacity, rising material costs, and longer lead times are complicating large-scale AI project delivery. MetaAge general manager Jason Shih said the imbalance between global AI hardware supply and demand will be difficult to resolve in the short term, even as enterprise AI investment continues and demand is expected to grow through 2027.

SoftBank Group (SBG) posted a JPY768 billion (approximately US$4.85 billion) net loss for the first half of fiscal 2026, from April to September 2026, its first loss in nearly two years after booking a US$3.7 billion impairment on a US data-center project and absorbing the impact of falling global AI valuations.

Meta has launched Muse Ember, its smallest model, to handle repetitive tasks at lower cost. Muse Ember features a 1-million-token context window, can generate up to 512,000 tokens per request, and supports text, images, audio, and video.

Strong iPhone sales and 5G upgrades, combined with continued revenue recognition from AI data center (AIDC), cloud, and enterprise information and communications technology (ICT) projects, drove sustained growth for Taiwan's three major telecom operators in October.

As AI servers and data centers expand, rising DRAM demand is driving up memory prices and supply pressures, making memory an increasingly important infrastructure cost. AMD has introduced AI-predictive memory management technology that tiers DRAM and NAND flash, forecasts application data-access needs, and moves data from Flash to DRAM before CPUs require it, reducing reliance on costly DRAM capacity.

Nvidia's board met on November 11 to weigh the future of Jensen Huang, the only chief executive the company has ever had. The stakes go beyond one career: the company at the center of the AI buildout has no publicly named successor, and the scandal is feeding a push in Washington to tighten control over its chips.

Moonshot AI reportedly accessed advanced Nvidia chips this summer, according to documents filed with Chinese regulators. The finding shows that advanced US chips continue to make their way into China despite Washington's export restrictions. It also reveals Moonshot's continued reliance on American chips even as Beijing seeks to bolster domestic chipmaking.

OpenAI's revenue reached US$10 billion in the third quarter of 2026, up nearly 50% quarter-on-quarter, driven by demand from enterprises and new model products. However, its adjusted operating loss widened to US$10.9 billion, while revenue remained below that of rival Anthropic.

Intel is expanding its CPU and GPU portfolio with the Intel GPU D-series, code-named Crescent Island, and next-generation Xeon processors, code-named Diamond Rapids, targeting large-model inference and enterprise high-performance orchestration workloads. The company outlined new GPU D-series technology at the 2026 Open Compute Project (OCP) Global Summit as it pursues server CPU upgrade opportunities driven by agentic AI.

Alibaba filed for a Hong Kong IPO for its chip design unit, T-Head, as a spin-off on November 10. The separate listing would let investors gain direct exposure to the group's chipmaking business as China's AI ecosystem seeks to reduce its reliance on US semiconductors through domestic alternatives.