Hanwha Semitech's supply of thermal compression bonding (TCB) equipment for high-bandwidth memory (HBM) to SK Hynix is drawing scrutiny after the South Korean equipment maker maintained a KRW100 billion (US$66.7 million) performance guarantee since early 2025, raising questions about whether its equipment has fully met the chipmaker's requirements.
The expansion of AI data centers is creating new business opportunities that are opening the door for upstarts to challenge industry leaders. A prime example is the four-level pulse amplitude modulation digital signal processing (PAM4 DSP) chip market for optical communications, which is shifting from a duopoly dominated by US leaders Marvell Technology and Broadcom to a more diversified vendor landscape. MediaTek's subsidiary Airoha is now striving to climb this competitive ladder.
SK Hynix has brought forward the ramp-up of its M15X DRAM facility in Cheongju, opening a second cleanroom and beginning equipment move-in about two months ahead of schedule amid rising demand for memory tied to artificial intelligence (AI), according to East Asia Daily.
Although it may appear that China's AI GPU development is behind the US by one or more generations, a recent Morgan Stanley report reveals the actual gap is smaller than widely perceived.
Alibaba's DAMO Academy has introduced two new RISC-V processors—the flagship XuanTie C950 and the energy-efficient C925—marking a significant step in pushing the open-source architecture into both high-performance AI and edge applications.
The Trump administration set up a US$250 million fund intended to grow out of its "Pax Silica" initiative launched in December 2025, aimed at supporting global supply chains for semiconductors, artificial intelligence (AI), and critical minerals, according to the New York Times. The fund represents the first step in a voluntary consortium that could attract up to US$1 trillion in investments from allies, including Singapore, the United Arab Emirates (UAE), Qatar, Sweden, Japan, South Korea, Israel, Britain, and Australia, with the US committing US$250 million.
Nvidia is expanding its role in artificial intelligence (AI) infrastructure by designing server racks that can accommodate processors from its competitors, according to a report by The Information, citing people familiar with the matter.
Tata Semiconductor Manufacturing's US$735 million loan package could accelerate India's domestic chip ambitions, but it also highlights execution and supply-chain risks for global technology supply chains. Lenders have insisted on brand continuity and majority control over the company while accepting leased land valued at zero as collateral, according to Mint and The Economic Times.
Huawei has officially launched its Ascend 950PR processor and introduced the Atlas 350 AI accelerator card equipped with this chip, marking the commercial debut of its next-generation inference computing platform. At Huawei China Partner Conference 2026, the company showcased the Ascend 950PR AI chip, highlighting that a single Atlas 350 card delivers up to 2.87x the compute power of Nvidia's H20, emphasizing its FP4 low-precision inference capabilities, a rare feature among AI accelerators in the Chinese market.
Taiwanese semiconductor and electronics suppliers face clearer incentives and growing opportunities in India following a budgetary push that prioritizes localized supply chains and higher-value-added segments. The developments could prompt more firms to reassess investment feasibility, though large foundry players have so far remained cautious.
Zeiss is significantly intensifying its investment in Taiwan, positioning the island as a premier global hub for its optical and semiconductor technologies. This strategic shift, marked by the appointment of Henry Cai as the new General Manager of Zeiss Taiwan, signals a move beyond product distribution toward deep-tier research and development collaboration.
More coverage
