AMD has introduced Helion, an open rack architecture designed for agentic AI workloads that could change how data centers allocate compute, power, and cooling. The system targets hyperscale and enterprise customers building AI agents that need continuous execution, tool use, and workflow coordination.

Unexpected behavior during Anthropic model testing—including models mistakenly submitting visa applications and sending false murder-case leads to police—has prompted the Trump administration to require AI companies to fulfill "national security obligations" by reporting related incidents to the government.

Microsoft has unveiled its latest in-house AI application-specific integrated circuit (ASIC), Maia 300, and accompanying server rack, with its compute chiplets manufactured using TSMC's advanced N2P process (an enhanced, extended version of the company's baseline 2nm, or N2). Supply-chain sources indicate that Lenovo, rather than Taiwanese manufacturers, will assemble the Maia 300 server racks.

Nanya Technology reported record profits in the third quarter of 2026, with president Pei-Ing Lee saying long-term agreements (LTAs) now cover approximately 50–70% of the company's production capacity. Some customers are seeking to extend contracts to three years or longer.

The AI buildout is still lifting Taiwan's server supply chain faster than almost any other industry, yet September's filings show the lift concentrating into fewer hands. The 49 listed Taiwanese companies tracked across the AI server chain reported combined September revenue of NT$2.95 trillion (US$92.35 billion), up 55.0% from a year earlier and 3.8% from August. Strip out Hon Hai Precision Industry, and the same group's September revenue fell 6.8% from August. For the first nine months, the group has booked NT$21.39 trillion, up 53.9% year-on-year.

As plans for the Terafab project begin to take shape, the emergence of a scaled chipmaking initiative increasingly reflects Elon Musk's broader ambition to push vertical AI integration across his expanding industrial stack.

China's leading artificial intelligence companies are gaining users and global recognition, but their commercial returns remain far behind those of US rivals. The growing disparity raises a question for China's AI industry: can low-cost, open-weight models generate sustainable profits, or will the greatest financial rewards flow to the platforms distributing them?

Electronic testing laboratory BTL reported September 2026 revenue of NT$87 million (US$2.73 million), up 10.2% year-over-year, as AI server testing projects at its second Guishan plant continued to drive growth. Revenue for the first three quarters reached NT$752 million, a 12.3% year-over-year increase, extending the company's double-digit growth trend since the start of 2026.

Capital expenditure by the top four US cloud service providers (CSPs) is projected to expand further in 2027. Meanwhile, nations such as France and Saudi Arabia are aggressively constructing sovereign AI infrastructure, ensuring that demand for AI hardware remains on a firm upward trajectory. In addition, shipments of next-generation AI servers powered by more capable Nvidia Vera Rubin chips will further fuel earnings growth across Taiwan's advanced semiconductor process nodes and advanced packaging sectors.

China's 15th Five-Year Plan and the Ministry of Industry and Information Technology's strong chip project are accelerating the domestic replacement of semiconductor equipment and materials amid intensifying US-China competition. Xiamen-based Jinai Tech, which integrates resources from both sides of the Taiwan Strait, is leveraging plasma, chemical vapor deposition (CVD), and atomic layer deposition (ALD) technologies to enter the advanced process equipment market, with its pilot production line scheduled to begin operations in mid-2027.

Driven by robust demand for AI, high-performance computing (HPC), and cloud data services, Taiwan's information and communications technology (ICT) production momentum has soared for three consecutive years. Meanwhile, mounting US and global debt levels are driving up real financing costs and tightening international financial conditions.