Global semi news — Korea, China, Taiwan, the US, and Japan. Government policy, export controls, capex moves, supply-chain shifts, and macro events. AI-classified and tagged with affected tickers. All headlines link back to the originating publisher.
Original: '이공계'에 미친 중국 지도부의 '뚝심'... 반도체에만 299조 쏟았다 - 한국일보
Korean press highlights that Beijing has channeled roughly KRW 299 trillion (~USD 220bn) into its semiconductor industry, underscoring Xi's long-term commitment to indigenous chip self-sufficiency. The scale reinforces concerns about state-backed Chinese competition in legacy/mature nodes, memory and equipment localization, with knock-on pricing and share pressure on Korean and Japanese incumbents.
Why it matters: Large-scale Chinese semi subsidy stories are sector-relevant for Korean/Japanese incumbents but the 299T figure is cumulative/recap rather than a fresh policy event, so impact is structural rather than near-term.
Original: 三菱電機、ローム、タツモ…エヌビディアも注目!「パワー半導体」でAIデータセンターの省エネ支える日本の最強技術企業【厳選8社】 | マーケットフォーカス - ダイヤモンド・オンライン
Diamond Online highlights eight Japanese power semiconductor specialists — including Mitsubishi Electric, Rohm and Tatsumo — whose SiC/GaN technology is gaining attention from NVIDIA and other AI data center builders seeking energy efficiency. The piece frames power semis as a key beneficiary of AI infrastructure buildout, where rising rack power density makes efficient power conversion a critical bottleneck.
Why it matters: Sector commentary on Japanese power semi suppliers benefiting from AI data center power infrastructure demand — relevant supplier-side angle but no specific catalyst or contract.
Open source articleOriginal: China AI Data Center Grid Locks Out Nvidia With $295 Billion Domestic Chip Mandate - Tech Times
China is reportedly mandating domestic AI chips across its $295B national data center grid buildout, effectively excluding Nvidia from a major share of Chinese AI infra demand. The policy shifts share toward Huawei Ascend and other local accelerators, while accelerating substitution risk for US AI chip exposure in China.
Why it matters: Specific $295B China domestic chip mandate that materially impacts Nvidia's China AI revenue and accelerates local accelerator substitution.
Open source articleOriginal: 엔비디아, Vera CPU 시스템 OpenAI·Anthropic·Oracle·SpaceXAI에 출하 개시
Nvidia has started delivering its next-generation Vera CPU systems to flagship AI customers including OpenAI, Anthropic, Oracle, and Elon Musk's SpaceXAI. The shipments mark the rollout of Nvidia's Vera Rubin platform, reinforcing its grip on AI infrastructure buildouts and signaling sustained hyperscaler capex into 2026.
Why it matters: Confirmed shipment of Nvidia's next-gen Vera CPU platform to top AI customers is a concrete product/demand event directly impacting NVDA and key supply-chain names.
Open source articleOriginal: 엔비디아, 과학용 슈퍼컴퓨팅에 에이전틱 AI 제안
Nvidia is promoting agentic AI workflows for scientific supercomputing, positioning its GPU stack and software ecosystem as the foundation for next-generation HPC research. The pitch extends Nvidia's AI infrastructure dominance into the scientific computing segment, reinforcing demand for high-end accelerators and associated HBM/networking components.
Why it matters: Sector-wide AI infrastructure theme extending Nvidia's GPU dominance into scientific HPC, with indirect read-through to HBM and networking suppliers.
Open source articleOriginal: 엔비디아 차세대 'Vera Rubin' 플랫폼, 모멘텀 가속
Yahoo Finance reports growing momentum behind Nvidia's next-generation Vera Rubin AI platform, the successor to Blackwell. The piece highlights continued demand for Nvidia's accelerator roadmap and downstream pull-through for HBM, advanced packaging, and AI infrastructure suppliers.
Why it matters: Roadmap momentum commentary on Nvidia's next-gen platform reinforces the AI infra theme and HBM/packaging supply chain pull, but contains no discrete new fact qualifying as 'high'.
Original: 엔비디아 차세대 Vera Rubin 플랫폼, 시장 모멘텀 가속
GuruFocus reports growing momentum behind Nvidia's next-gen Vera Rubin AI platform, the successor to Blackwell. Rising customer interest reinforces the AI infrastructure capex cycle and benefits HBM, advanced packaging, and networking suppliers across the Nvidia stack.
Why it matters: Momentum commentary on a previously announced roadmap product reinforces the AI infra theme rather than delivering a discrete new event.
Original: 엔비디아 차세대 Vera Rubin 플랫폼, 모멘텀 가속
GuruFocus reports growing momentum behind Nvidia's next-gen Vera Rubin AI platform, signaling sustained demand for advanced GPU architectures. The ramp implies continued pull-through for HBM suppliers, CoWoS packaging capacity, and networking partners in the AI infrastructure stack.
Why it matters: Repackaged commentary on Nvidia's next-gen platform momentum — directionally bullish for the AI infra supply chain but lacks a hard new datapoint.
Original: 'AI 심장' HBM 삼국지…마이크론, 앤스로픽과 동맹 선언 - 리드경제
Micron announced a strategic alliance with Anthropic to supply HBM for AI workloads, intensifying the three-way HBM competition with SK Hynix and Samsung. The deal signals Micron's accelerating push into the AI memory tier where SK Hynix currently leads via NVIDIA, raising share-loss risk for Korean incumbents if Anthropic-linked demand scales.
Why it matters: A named hyperscaler-memory alliance directly targeting the HBM tier where SK Hynix and Samsung dominate is a near-term competitive event for Korean majors.
Open source articleOriginal: Microsoft plans 2GW data center campus in Pecos, Texas
Microsoft is planning a 2GW data center campus in Pecos, Texas, with Chevron signed on as the energy supply partner. The scale signals a major incremental AI capex commitment and reinforces the power-constrained DC buildout thesis, with read-through to gas turbine/power-infra suppliers and AI accelerator/memory demand.
Why it matters: Hyperscaler DC capex announcement with specific 2GW capacity figure and named power partner signals sustained AI infra demand, but no direct chip-supplier impact disclosed.
Jul 14, 2026 close · day-over-day
NBIS
$194
-7.80%