Huawei expects autonomous AI agents to become the dominant source of AI traffic over the next decade, creating a sharp increase in demand for computing, networking, storage and power infrastructure.
ByteDance has completed a US$290 million funding round for Anew Labs, its AI drug-discovery business spun off from the Chinese technology group, valuing the company at US$1.5 billion, according to Reuters. ByteDance will retain a 56% stake following the fundraising.
Meta Platforms is preparing a new pair of smart glasses without a camera, a move that could address a key social barrier facing wearable devices as the market expands rapidly.
Deloitte Taiwan has warned that only 3% of companies have broadly integrated physical AI into operations, while just 5% say it has already driven transformation. The firm said adoption remains slow because physical AI involves compliance and safety issues, and its growth pace is nowhere near that of agentic AI or generative AI.
BizLink chairman Roger Liang said customer demand, market access, supply chain resilience, and local service needs have driven the company's overseas expansion as geopolitics, tariffs, and localization pressures reshape global manufacturing. He said every site, venture, and acquisition must be judged on controllable risk, reasonable returns, and long-term growth.
As global AI investment continues to expand and US-China geopolitical pressure intensifies, how Taiwan and South Korea can preserve their semiconductor and manufacturing advantages has become a focal point. DIGITIMES and IC975 chairman Colley Hwang said the two sides should look for complementary strengths, work together to help Vietnam, Malaysia, and other countries develop, and first increase corporate exchanges to build a positive environment for cooperation.
DIGITIMES Intelligence said Google used SEMICON Taiwan 2026 to outline the direction of its AI infrastructure strategy, focusing not only on next-generation TPU products but also on how rising AI compute demand is accelerating TPU iteration and deployment.
Wiwynn held the grand opening for its first US factory on September 14 local time, choosing El Paso, Texas, over larger cities such as Dallas and Houston. Chief manufacturing officer Robin Wang said the site gives the company the best support from its Ciudad Juarez, Mexico plant, which is less than an hour away.
Samsung Electronics' in-house Exynos application processors (APs) are steadily regaining lost ground in the global smartphone market. Samsung's share of the global smartphone AP market rose to 9% in the second quarter of 2026, up 2 percentage points from the previous quarter and 3 percentage points from a year earlier, marking a second straight quarterly increase.
Altera, the programmable-chip maker backed by Silver Lake and Intel, has confidentially filed for an initial public offering (IPO) in the US, positioning the company to tap renewed investor demand for semiconductor and AI-related businesses.
Where will China's electronics manufacturing industry find growth over the next five years? The answer is taking shape in the 15th Five-Year Plan for the sector, released by the Ministry of Industry and Information Technology (MIIT) and the National Development and Reform Commission (NDRC): industry revenue is targeted to exceed CNY30 trillion (US$4.47 trillion) by 2030.
Taiwan's AI server supply chain continued to show strong year-over-year growth in August, although performance varied significantly across baseboard management controller (BMC), cable/connector, and server-case suppliers.
DIGITIMES observed that Nvidia's newly announced NVHBM is not a brand-new memory technology, but rather an extension of existing custom HBM (cHBM). It is a managed cHBM design that moves the memory controller, which originally sat on the XPU, into the HBM logic base die, and pairs it with custom physical-layer (PHY) and die-to-die (D2D) interfaces. In strategic terms, Nvidia is extending its moat from GPUs to the memory architecture of custom ASICs, and is trying to define how future custom AI chips connect to memory, hook into scale-up networks, and come together as complete AI racks.
