At a regular briefing Foreign Ministry spokesman Guo Jiakun said China has noted recent US industry calls to slow frontier AI development. He said China pursues development and security in parallel and attaches high importance to AI’s intrinsic and derivative risks. Relevant Chinese agencies yesterday issued the AI Security Governance Framework 3.0, which updates AI safety risk classifications and recommends technical responses and comprehensive governance measures. Guo said China supports using

China’s Ministry of Industry and Information Technology and National Development and Reform Commission issued a joint electronic information manufacturing plan calling for accelerated development of advanced storage technologies and compute infrastructure. Policy directs promotion of storage‑centric computing and compute‑by‑storage, upgrades to distributed and hyperconverged storage, and the establishment of tiered data‑storage architectures and protocols. It targets breakthroughs in intelligent tiered storage, data sparsification, key‑value caching and the R&D of high‑bandwidth flash, high‑bandwidth memory (HBM), ferroelectric, resistive (ReRAM) and magnetoelectric memories. On facilities and applications, the plan backs hyperscale intelligent‑compute architecture to support AI clusters at the scale of hundreds of thousands of accelerator cards, a full‑chain buildout of space‑based computing with deeper compute‑power co‑design, strengthened distributed edge computing architectures, higher‑performance compilers/databases/middleware, and work on domain‑specific accelerators and large‑scale parallel compute bottlenecks to lift Z‑class supercomputer performance.

波蘭央行行長格拉平斯基：利率可能在2026年底前保持不變；不能排除2027年第一季度加息的可能性。