Advancements in semiconductor technology, GPU development, and data center infrastructure that power AI computing capabilities.
Researchers at the University of Arizona have developed a new optical switching technology, the PQT-HOS, which operates 1,000 times faster than electrical switches and uses only one-thousandth of the energy, potentially reducing data center power consumption.
SoftBank founder Masayoshi Son is in discussions with French President Emmanuel Macron over a potential $100 billion investment to build artificial intelligence data centers in France, according to Bloomberg. The project could be announced at the upcoming Choose France Summit.
3M has joined a coalition of major technology companies to develop open standards for expanded beam optical connectivity used in AI data centers. The group aims to establish interoperable specifications to accelerate high-performance optical interconnect deployment.
Nscale has obtained $790 million in financing from a group of Nordic and European banks to support the expansion of its AI data center in Narvik, Norway. The deal includes an option for an additional $790 million to fund a further 115 MW of capacity.
atNorth has joined Nordic Compass, a new alliance of over 25 Nordic organizations focused on strengthening competitiveness, resilience, and industrial innovation across the region.
Cyient Semiconductors has introduced India's first gallium nitride power IC family, developed under license from Navitas Semiconductor. The seven-device portfolio targets applications in AI data centers, telecom, consumer power, and e-mobility, with sampling expected to begin in June 2026.
VIVIFY Technology founder Jason Herring presented the company's closed loop hydrogen power platform during a Florida press conference focused on AI and data center infrastructure, highlighting its potential to reduce grid dependency and water use.
AZIO AI has received an order from Envirotech Vehicles to expand modular AI data center infrastructure in South Texas, adding about 5 megawatts of compute capacity powered by behind-the-meter natural gas generation.
Everpure has introduced new capabilities for Red Hat OpenShift users, integrating Portworx data management directly into the OpenShift console to simplify storage, protection, and disaster recovery for AI workloads, containers, and virtual machines.
Zyphra has announced 15 megawatts of AMD Instinct MI355X GPU capacity through Zyphra Cloud, adding new options for AI workload deployment on bare-metal AMD infrastructure.
At the 2026 European Economic Congress, MICROIP announced a partnership with Poland to establish a resilient Edge AI and ASIC supply chain, introducing its 'Software Driven Hardware' strategy to address AI deployment challenges.
FotoNation and SEMIFIVE have announced a collaboration to develop the TriSilica perceptual AI chip family, using Samsung Foundry’s 8nm process. The initial TS-210 chip is planned for a Multi-Project Wafer shuttle by the end of the year.
Target Hospitality reported $72.8 million in revenue and a net loss of $13 million for the first quarter of 2026. The company announced over $1.4 billion in new multi-year contracts supporting AI infrastructure and power generation projects across Texas.
The Shanghai Lin-gang Underwater Data Center, the first wind-powered underwater data facility in the world, has officially started operation off Nanhuizui in Shanghai, using offshore wind and seawater cooling to achieve a power usage effectiveness below 1.15.
Sterlite Technologies Ltd announced plans to invest up to $100 million in the United States to expand its manufacturing capacity for optical fiber connectivity solutions supporting AI data centers and telecom customers.
MountAIn has announced a partnership with Alif Semiconductor to deliver a cloud-grade computer vision stack optimized for low-power processors, enabling AI model deployment directly on the edge.
Sivers Semiconductors has announced a $1.5 million development partnership with Tachyon Networks to create a 60 GHz mmWave transceiver aimed at expanding fixed wireless access capabilities.
OpenAI's plan to produce custom AI chips with Broadcom, aimed at reducing reliance on Nvidia, is facing an $18 billion financing hurdle as Broadcom seeks a purchase commitment from Microsoft for 40% of the chips.
Hut 8 has signed a 15-year lease valued at $9.8 billion for its Beacon Point data center campus in Texas. The facility, designed to support large-scale AI operations, could reach a value of $25.1 billion if renewal options are exercised, according to Bloomberg.
Verda and Compal Electronics have formed a strategic partnership in which Compal will supply GPU server systems to accelerate Verda's AI cloud infrastructure across Europe and the Asia-Pacific region.
Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.
