Arm has brought on Rami Sinno, the former Amazon Web Services (AWS) executive who helped lead development of the Trainium and Inferentia processors, along with senior engineers from Hewlett Packard Enterprise and Qualcomm. The hires underscore Arm’s push to expand beyond processor blueprints into cloud chip and system-level design.

As the US government seriously considers taking a stake in Intel, the company faces more than just a simple transaction; it has become a strategic industry bet. Once the semiconductor leader, Intel’s lagging advanced process technology has gradually eroded its ability to compete with TSMC and Samsung Electronics at the 7nm, 5nm, and even upcoming 2nm nodes. Meanwhile, construction delays at the Ohio wafer fab have hindered the US government’s carefully crafted blueprint for domestic advanced manufacturing from materializing.

The global race to build ever-larger AI models is intensifying, and the battle is no longer confined to Nvidia’s powerful GPUs. Another crucial, though less visible, component has emerged as the bottleneck: high-bandwidth memory (HBM).

Alibaba Group’s AI code-writing model, Qwen3-Coder, has become the world’s second most used AI coding tool within a month of its July 23 launch. It holds a significant 20% share of global usage on the OpenRouter platform, behind Anthropic’s Claude Sonnet 4 at 31%.