At its investor conference on November 27, 2025, Gigastorage Corporation shared details regarding its strategic shift from its early photovoltaic (PV) business toward AI computing, cooling, and personal health micro-sensing. In the HPC and AI fields, Gigastorage has invested in direct forming technology (DFT). This technique leverages the high hardness of diamond abrasives to directly machine deep, tall two-dimensional fin structures onto heat-dissipating materials, improving thermal resistance performance by approximately 20%. It overcomes the technical bottlenecks of traditional cooling solutions, allowing fins to be directly integrated into GPU modules while precisely controlling the base thickness below 0.2mm, targeting top-tier applications such as HPC and vertical AI cooling.

The US Genesis Mission, a major AI infrastructure project, has boosted Taiwan's AI chip supply chain, as all global AI inference chips come from Taiwanese firms. With 76% of TSMC's revenue from the US, improved US-China relations and a lifted Chinese import ban could further surge Taiwan's AI orders.

Swancor Advanced Materials, the STAR Market–listed maker of high-performance composites and corrosion-resistant materials, has entered a new phase after naming Peng Zhihui — the Bilibili tech star "Zhi Hui Jun" — as chairman. All nine directors unanimously approved the move at the board's first meeting on November 25, 2024, in compliance with Company Law and STAR Market rules, according to ICsmart and 36Kr.

Recent market rumors indicate that Meta is negotiating to deploy Google's seventh-generation TPU "Ironwood" in its data centers by 2027, with procurement potentially reaching tens of billions of US dollars. This move signals confidence that Google will become a leading ASIC chip supplier, challenging Nvidia's dominance and securing the second spot in the AI accelerator market.

Acer Group subsidiary Highpoint Service Network (HSN) is charting its next phase of expansion after recording three consecutive years of double-digit growth, driven by a "multi-country, multi-product, multi-brand" strategy. Chairman Andrew Hou said the company is evaluating market entry into India or China and aims to maintain double-digit growth through 2026.

Kneron introduced a new generation of AI chips on November 26, 2024, as the company positions itself as a full-stack AI infrastructure provider. The launch featured the flagship KL1140 and included guest attendees such as Macronix chairman Miin Wu and Etron Technology chairman Nicky Lu. Founder and CEO Albert Liu also outlined the company's three-year roadmap covering high, mid and entry-level processors, marking what he described as a milestone in completing Kneron's full compute portfolio.

Explosive AI demand is accelerating a historic global memory chip shortage, with major cloud service providers (CSPs) securing multi-year long-term agreements (LTAs) to guarantee supply through 2027 and 2028. Industry sources reveal that nearly all memory production capacity for 2026 has been pre-booked, confirming an unrelenting shortage throughout the year.

The humanoid robot sector in China is experiencing rapid growth in production capacity, despite unclear large-scale demand, as companies prepare for anticipated commercialization. This comes amid delays in mass production efforts by Tesla Inc. in the US, highlighting differing dynamics between global markets.

As 2025 wraps up, MiTAC Holdings president Billy Ho has stated that it has been a year of changes and challenges. He noted that 2026 will still have obstacles, including geopolitics, national policies, high levels of government debt, and weak global economic growth. Despite these challenges, he expressed confidence that growth in 2026 will not be an issue for MiTAC, with the second half expected to perform better than the first half.

In the competitive realm of artificial intelligence computing power, Nvidia currently leads the market, but Google is emerging as a formidable contender with its self-developed TPUs. According to CNBC, Google has collaborated extensively with Broadcom since 2016 to design and manufacture its AI chips. This partnership has progressed into the seventh generation of TPUs, which power Google's internal AI infrastructure and rival Nvidia's GPUs in AI workloads.

China's AI chip startup Zhonghao Xinying has introduced its own tensor processing unit (TPU), a major step in the country's "de-Americanisation" drive as US export curbs continue to block access to Nvidia's high-end GPUs.