At the OCP Global Summit, Intel introduced GPU D-series accelerators aimed at data-center inference and agentic AI workloads. The company said the product was previously known by the codename Crescent Island and customer sampling is underway.
The USFP optical module format targets higher connectivity capacity in AI factories. The companies specify 400 Gb/s lanes and up to 12.8 Tb/s total capacity.
gFiber, Qualcomm and Gemtek unveiled a Wi-Fi 8 home gateway designed for gFiber fiber customers in the United States. The joint announcement describes hardware combining multi-gigabit fiber access, wireless networking and on-device AI processing.
Google Research published one roundup of ten research and tool highlights for Research@2026 on October 14. Topics span geospatial AI, scientific workflows, chip development and health research.
The Cabinet Office, AI Security Institute and National Cyber Security Centre said two existing pilots had identified and addressed more than 1,000 potential vulnerabilities. New work includes evaluating advanced AI and preparing for incidents involving autonomous agents.
Researchers at KTH Royal Institute of Technology introduced GraphTok, a system that tokenizes large language model requests on a GPU. In a Dynamo frontend experiment, a real H100 NVL performed tokenization while backend GPU workers were simulated.
Researchers affiliated with Hong Kong Polytechnic University and other institutions in China modeled optical circuit switching alongside network scheduling to shift AI data-center training toward renewable-energy windows. In a controlled five-center scenario, modeled grid-purchase cost was 12.50% lower and local renewable-energy use rose by 2.724 percentage points.