Nvidia said in its fiscal 2027 second-quarter results on the 27th that it sold a small number of H200 artificial intelligence (AI) processors to Chinese customers last quarter, marking its first renewed AI chip sales to China since shipping about US$60 million of H20 chips in early 2025. But shipments fell short of the total approved by US President Donald Trump and licenses cleared by the US government in January, amid opposition from Beijing.

Nvidia closed its second fiscal quarter with US$31.6 billion of inventory, but the more consequential change is what that inventory now consists of. Work in process and raw materials together account for 78% of the balance, against 42% a year earlier. The mix shows a company committing to scarce memory well ahead of the Vera Rubin ramp, and absorbing the cost several quarters before customer price increases land.

Nvidia's data center business now splits almost evenly between hyperscalers and everyone else — US$48.7 billion against US$40.3 billion in the July quarter. The obvious reading is that the customer base is broadening. The sequence says something more specific.

The recent disclosure of SpaceX's partnership with Nvidia to deploy Vera CPUs as part of the architecture behind SpaceXAI's Starmind AI satellite was an outright showcase of where the two tech giants are decisively headed in the agentic AI era, as their collaboration deepens into one huge, interconnected ecosystem of orbital compute.

Google and MediaTek are moving toward advanced packaging diversification, as Intel's embedded multi-die interconnect bridge (EMIB) gains traction after Google's TPU was confirmed to adopt the packaging technology. The shift is fueling expectations that more advanced packaging orders could move to Intel, while TSMC's chip-on-wafer-on-substrate (CoWoS) remains tight on capacity.

Weltrend Semiconductor said first-half 2026 revenue rose 28% from a year earlier to a record for the period, with growth driven almost entirely by cooling demand tied to AI data center investment.

Xiaomi is expanding Xring from a smartphone chip program into a broader AI computing platform spanning consumer devices, vehicles, and edge inference, with three processors assigned to distinct workloads.

Amazon Web Services (AWS) and Nvidia are expanding their collaboration to add millions of GPUs and new infrastructure for agentic and physical AI. The move signals rising global demand for advanced computing, with implications for cloud customers, governments, and robotics developers seeking faster, more secure access to AI tools worldwide.

For the first time in this cycle, Nvidia has told investors where its gross margin will bottom out and how far short of demand its supply will fall - two admissions that shift the question from whether the AI buildout continues to who absorbs its rising input costs. Management guided gross margin down from 75.0% to a trough of 71-72% by the fourth quarter, directly blamed memory pricing, and capped fiscal 2028 revenue growth at about 70% against demand it put at nearer 100%. Neither disclosure is a demand warning. Both say the company is now managing a shortage rather than a market.

Growth has stopped being the interesting question at Nvidia. Revenue for the quarter ended July 26, 2026 was US$96.2 billion, up 106% from a year earlier, and Data Center accounted for US$89.0 billion of it, up 117%. Gross margin held at 75.0% even as the Vera Rubin platform moved into volume—unusual, because a new rack-scale architecture normally costs a quarter or two of margin while yields settle. On the operating line, the quarter was close to flawless.

Nvidia has put a figure on its exposure to the AI labs it helps fund. On its second-quarter earnings call, chief financial officer Colette Kress said the company expects "demand from the AI labs for which we expect to leverage our balance sheet to contribute toward roughly a quarter of our business next year." That single sentence reframes the financing question. The issue is no longer whether Nvidia supports its customers' balance sheets, but that a quarter of forecast revenue now depends on structures the company itself has arranged.

The sharpest question on Nvidia's second-quarter call came a Bank of America analyst, who noted that OpenAI and Anthropic - both major beneficiaries of Nvidia's ecosystem investment - are designing their own chips, and that OpenAI had, days earlier, claimed its Jalapeno part outperforms Blackwell.

Nvidia has guided a full year in advance for the first time, and the number it chose is not a demand forecast. Chief financial officer Colette Kress told the company's second-quarter earnings call that Nvidia expects to grow revenue by approximately 70% in fiscal 2028, then immediately qualified it: "This is a supply-constrained outlook." Customer forecasts, she said, "point to our growth doubling next year." The 30-point gap between what buyers want and what Nvidia has committed to ship is now the central fact about the company's next four quarters, and it is what the Vera Rubin, Groq, and Vera CPU ramps exist to close.

Apple took pre-orders on August 25 for a new Mac mini and Mac Studio. Both will ship on September 22. Prices of the Mac mini start at US$899 with the M6 chip, or $1,699 with the M5 Pro. Mac Studio starts at $2,499 with M5 Max and $5,499 with M5 Ultra. The 512GB memory version comes in late October.

Nvidia will release its fiscal second-quarter 2027 financial results on August 26, US time. Given Nvidia's performance pattern in recent quarters, Wall Street has come to expect a "beat and raise" outcome. Ahead of the print, multiple investment firms released preview reports focusing on whether the Rubin platform can smoothly succeed Blackwell and whether AI demand continues to significantly outstrip supply.

Enflame's planned US$893 million STAR Market IPO, examined in Part 1 of this series, highlights both the opportunity and the financial strain facing China's emerging AI chip suppliers, but profitability is only one part of the competition.

Enflame Technology is preparing to become the latest of China's leading AI chip developers to tap public markets, with a CNY6 billion (approx. US$893 million) Shanghai STAR Market listing that puts its rapid revenue growth, persistent losses and heavy reliance on Tencent under closer scrutiny.

Google's expanding list of TPU partners has sparked widespread industry discussion, centering on whether current IC design suppliers Broadcom and MediaTek will see their orders impacted.

India's semiconductor expansion is beginning to expose a shortage of engineers with experience in production testing, a capability that will become increasingly important as new OSAT plants and fabs move toward manufacturing, executives at Emerson's NI test and measurement business said.

Nvidia has introduced Jetson Orin Nano 2, a compact robotics computer aimed at entry-level edge AI, with implications for developers building autonomous machines worldwide. The system promises faster, more efficient on-device intelligence for robots, drones, and vision systems as physical AI adoption expands across industries and markets.

OpenAI said its first custom inference chip, Jalapeño, delivered faster responses and better power efficiency than competing systems in tests across several large language models. The findings could matter for global users, as cheaper, lower-latency AI infrastructure may help expand access, improve reliability, and support more capable agents worldwide.

Xiaomi officially unveiled its next-generation in-house system-on-chip (SoC), the Xring O3, at its Xring Chip Technology Briefing on August 24, continuing to position the silicon as an AI flagship SoC.

Samsung Electronics and Intel are pushing into each other's core businesses as AI semiconductor competition shifts from individual chips to full-system integration. South Korean reports cited by Counterpoint Research said Samsung is using high-bandwidth memory (HBM) to strengthen its foundry business, while Intel is expanding from logic chips and foundry services into memory through advanced packaging and system-level technologies.

As robotaxi services expand beyond individual cities into wider regions, the growing appeal of AI-assisted driving and the technological vision that AV developers are eager to realize have raised industry expectations for the economic viability of full autonomy.

Nexperia China said it has rebuilt a domestic 12-inch wafer supply chain after 330 days of overseas wafer supply disruptions, unveiling the new platform at a press conference outlining its new direction. The company said complete China-based manufacturing supply chains have now been established for metal-oxide-semiconductor field-effect transistor (MOSFET) and logic IC products.