nvidia

TSMC's Compact Universal Photonic Engine enters volume production in 2026, with NVIDIA's Spectrum-X Photonics switches and AMD's new $280 million R&D hub positioning optical interconnects as the successor to copper wiring in AI infrastructure.

NVIDIA released the Omniverse DSX Blueprint at GTC 2026, a framework for building physically accurate digital twins of AI factories that unifies power, cooling, and network simulation.

Linux 7.0, expected in mid-April 2026, formally ends the three-year Rust experiment in the kernel. The decision, reached at the 2025 Maintainers Summit in Tokyo, promotes Rust from experimental status to a core kernel language alongside C and assembly.

TSMC's US footprint could triple from $165 billion to $465 billion under a proposed tariff deal, with four Arizona fabs already fully booked before construction is complete.

Chinese AV firm WeRide and Southeast Asian superapp Grab begin public autonomous rides in Singapore's Punggol district, marking the city-state's first driverless service open to everyday residents.

Nvidia takes a $2 billion stake in Marvell Technology, integrating one of the semiconductor industry's top custom ASIC designers into its NVLink Fusion interconnect platform.

The French AI startup raised $830M from seven banks to install 13,800 Nvidia GB300 GPUs at a 44 MW facility south of Paris, as it moves from renting cloud capacity to owning infrastructure.

Nvidia open-sources a 30B mixture-of-experts model that activates only 3B parameters yet matches DeepSeek's 671B model on olympiad-level math and coding benchmarks.

AMD's Instinct MI400 family debuts three CDNA 5 accelerators on TSMC 2nm alongside the 72-GPU Helios rack, but a production timeline dispute clouds the second-half 2026 launch.

Fanuc is integrating Nvidia's simulation and inference platforms across its industrial robot portfolio while committing $90 million to a new 840,000-square-foot Michigan facility, positioning the world's largest robot maker to meet surging North American demand for AI-driven factory automation.

NVIDIA transferred its Dynamic Resource Allocation driver for GPUs to community ownership under the Kubernetes project at KubeCon Europe 2026, while Google open-sourced the GKE Cluster Autoscaler and Broadcom donated Velero to the CNCF Sandbox.

The five largest US cloud providers have committed up to 690 billion dollars in 2026 capex, nearly doubling 2025 levels, as inference workloads surpass training to consume 60 to 70 percent of AI compute demand.