Oracle's US$165 billion data center project in New Mexico is running into higher-than-expected costs due to local concerns about its energy requirements. This case is one of several showing how the build-out of AI infrastructure in the US has become a politically charged issue at the local level due to worries over electricity needs and other issues.
Meta Platforms is in early talks to lease computing power to Anthropic in a potential deal worth up to US$10 billion over two years, as the social media company considers generating revenue from its expanding AI infrastructure.
AI development is driving larger data transfers and higher GPU efficiency demands, pushing memory toward customization and prompting South Korean industry watchers to call for a shift to a "memory foundry" model. Sungkyunkwan University professor Seokjoon Kwon said at the Nano Korea 2026 forum that memory makers must move from mass production to order-driven design tailored to customer needs.
SiPearl has powered on and begun validating Rhea1, its first-generation server CPU designed in Europe. Its next challenge is convincing Taiwan's server manufacturers to turn the chip into systems that data centers can order and deploy.
The surge in AI and high-performance computing (HPC) technologies continues to reshape the semiconductor landscape, as demonstrated by the June 2026 financial performances of key supply chain players in Taiwan. Within this thriving ecosystem, IC testing and design services have emerged as pivotal backbones ensuring the physical viability and operational efficiency of next-generation silicon.
Nvidia has introduced new Thor-based modules aimed at powering robots and autonomous machines as they move into broader commercial use. The launch reflects rising global demand for compact, efficient AI hardware that can run advanced foundation models on-device, helping businesses deploy smarter systems with lower latency and reduced reliance on the cloud.
A European industry group, the Cloud Infrastructure Services Providers in Europe (CISPE), has joined four other trade organizations in calling for the European Commission to impose interim measures while it processes an antitrust case against Broadcom. The case concerns recent licensing changes made by the chip designer on the virtualization platform VMware.
Nvidia co-founder and chief executive Jensen Huang used a developer event in Tokyo on July 15 to reject reports that manufacturing problems could delay its next-generation AI accelerator systems, telling reporters the claims were "not true" and that "Vera Rubin is already in production. Giant amounts of production incoming."
Nvidia has laid out a sweeping expansion of its Japanese footprint. The company is moving beyond one-off supercomputer wins to embed its Blackwell-generation chips and software across the country's research labs, banks, hospitals, factories, and automakers. The breadth signals that Japan is being positioned as a full "AI ecosystem" for Nvidia, not a single-sector customer. It's a hedge that spreads the company's growth across sovereign science, industrial automation, and physical AI, even as questions mount over chip pricing and supply.
Taiwan's AI server component suppliers generally maintained healthy revenue momentum in June, with power supply, thermal solution and baseboard management controller (BMC) vendors continuing to benefit from strong AI infrastructure demand. Optical module suppliers, meanwhile, delivered a more mixed performance, with several companies posting triple-digit or strong double-digit annual growth while others remained under pressure.
