JCET Group is accelerating its AI chip packaging expansion with a CNY7.8 billion (US$1.1 billion) investment in a new advanced packaging and testing plant in Shanghai Lingang, strengthening capacity for AI computing, high-performance chips, high-density memory and automotive electronics.
Global helium supply is under renewed strain. Nippon Sanso, Japan's largest industrial gas supplier, announced it will raise prices across its helium product line by an average of more than 30% starting July 2026, citing persistent tightness in global supply driven in part by rising geopolitical risks in the Middle East.
As demand for AI chips surges, the battle for foundry orders is heating up. South Korean industry sources say the next 2-3 years will be a critical period for Samsung Electronics' foundry business, with its 2028 turnaround target hinging on stable operations at the Taylor plant and landing major orders.
Rising demand for AI server builds is tightening the global supply of high-end MLCCs, driving steep price increases in China's electronics distribution market and exposing how quickly infrastructure demand can affect component costs worldwide. The rally underscores a structural mismatch that could keep pressure on buyers across technology supply chains for years, according to industry sources.
ByteDance's reported plan to purchase at least 50,000 AI inference chips from Shanghai-based GPU developer Iluvatar CoreX could become one of the most significant commercial wins yet for China's domestic AI chip industry.
IC design firms are scrambling to expand their partner networks as tight packaging and testing capacity remain the most pressing bottleneck in the supply chain. Companies in Taiwan, Europe, and the US say the shortage is worse than expected, pushing packaging, testing, and related materials into broader supply constraints.
At Nvidia's June 25 annual shareholder meeting, CEO Jensen Huang declared that national security "comes first" wherever it conflicts with commercial opportunity, pledging full compliance with US export controls while casting the chipmaker as a core pillar of America's AI and semiconductor industrial base.
MediaTek is reportedly strengthening its partnership with Google in ASICs, a move that could increase the scale of future orders and carry implications for AI infrastructure worldwide. Market talk suggests the company may build an upgraded triggerfish product for Google, underscoring how global chipmakers are vying for influence in TPU development.
Qualcomm Technologies is expanding its partnership with Hugging Face to bring open AI tools from devices to cloud infrastructure, a move that could affect developers and enterprises worldwide. The collaboration aims to simplify AI deployment across the compute continuum while enabling faster, more flexible, and more scalable hybrid AI applications.
SK Hynix said on June 24 its board approved a plan to issue new shares backing American Depositary Receipts on the Nasdaq Global Select Market, targeting up to KRW45.45 trillion (approx. US$29.43 billion) in proceeds for semiconductor facility investment.
OpenAI on June 24 unveiled "Jalapeño," its first in-house AI accelerator, co-developed with Broadcom and manufactured by TSMC — the clearest sign yet that leading model developers want to design their own chips and broaden a supply chain long dominated by Nvidia.
More coverage
