Huawei argues its new architecture could achieve performance on par with 1.4-nanometer node density by 2031, while China currently manufactures chips around 7 nanometers and TSMC progresses toward 2nm and 1.4nm production.

The analysis notes risks around AI development, including potential distillation attacks that copy American models, and sketches possible futures for 2028 shaped by policy responses.

Move away from Moore’s Law to a Tau Scaling Law that prioritizes data transfer speed and 3D stacking to overcome shrinking transistor limits.

Broader AI hardware activity is evident as Microsoft updates Copilot integration in Windows 11 and HP rolls out ZBook AI chips and Blackwell GPUs, signaling a thriving ecosystem.

The strategy unfolds amid US export controls on lithography tech, pushing Huawei to seek non-Nvidia hardware alternatives for AI workloads inside China.

Industry observers flag heat management, cost, power use, and integration as the main obstacles to scalable adoption of advanced architectures.

HiSilicon plans to test LogicFolding in fall 2026 with the Kirin line, aiming to extend to Ascend AI chips and large data-center clusters by 2030.

State media frame competition as a matter of moderation and mutual advancement, with high-level AI governance talks noted between the U.S. and China.

Huawei introduces LogicFolding, a 3D stacked design meant to boost performance by stacking circuits and reducing reliance on cutting-edge lithography.

Market commentary highlights Nvidia, AMD, and Huawei’s breakthroughs within a tense tech race, with China remaining a large market for chipmakers.

Huawei formalizes Tau Scaling Law as a guide to chip development beyond Moore’s Law, aiming for 1.4nm-equivalent density by 2031.

Analyses warn that the next two years will determine whether democracies or authoritarians lead in AI, stressing control gaps and vulnerabilities in current safeguards.