A draft State Department memo obtained by Reuters on Aug. 14 reveals a blunt diplomatic overture: the United States wants the 35 nations that signed its AI Opportunity Statement to pick a side in the growing technology rivalry with China. The internal document, referenced by a senior U.S. official, tells the recipients that “to be part of everything is to be part of nothing.” It frames the Pax Silica declaration not as a membership subscription but as a binding commitment to a shared vision on artificial intelligence, semiconductors and critical minerals.

Pax Silica, launched last year, aims to secure supply chains for three strategic pillars—AI models, chips and the minerals that power advanced hardware. Roughly two dozen countries, including Japan, Australia and South Korea, have already joined the core group. The broader AI Opportunity Statement, a symbolic pledge of “common purpose,” now encompasses 35 signatories. While the framework is non‑binding, the draft letter seeks to close that gap by demanding an explicit choice.

Kazakhstan sits at the center of the drama. The Central Asian nation joined Pax Silica in June, a move highlighted by Under Secretary Jacob Helberg as the first accession from the region. Shortly after, Kazakhstan also signed China’s World Artificial Intelligence Cooperation Organization, the only known country to belong to both initiatives. Washington officials say the dual membership set off alarm bells, fearing that the country could hedge between rival blocs and undermine the very supply‑chain security Pax Silica intends to guarantee.

China’s competing effort, announced by President Xi Jinping in July, promotes open‑weight AI models that can be downloaded freely, contrasting with Pax Silica’s emphasis on shared investment opportunities and coordinated export controls. Though framed as an open‑access platform, recent reports suggest Beijing is weighing restrictions on overseas access to its leading models, indicating a more guarded approach than the public pitch.

The U.S. memo never mentions China by name, instead referring to “duplicative initiatives whose expectations conflict with our own.” A source familiar with the draft said the language is meant to make clear that countries cannot simultaneously align with both ecosystems. The official added, “You can’t have it both ways.”

Timing appears strategic. Chinese open‑weight models have narrowed the gap with proprietary systems from OpenAI and Anthropic, with industry insiders noting China now accounts for roughly 41% of global model downloads. At the same time, American chipmakers such as Nvidia are grappling with export restrictions, and the Commerce Department is probing how China accesses Nvidia hardware it cannot purchase directly. Treasury officials have even threatened sanctions on Chinese AI firms.

Europe is conspicuously absent from the reported member list, though the article does not publish the full roster. The European Union has pursued a third‑way strategy, emphasizing local content rules for hardware and sovereign compute capabilities, suggesting a reluctance to fully align with either side.

Analysts warn the letter could backfire. China analyst Rui Ma argued on X that coercing nations into a binary choice may drive them toward greater self‑reliance or push them toward the open‑access model China offers. Other technology developers echo the sentiment, urging countries to build independent AI ecosystems rather than depend on either bloc.

Three variables will determine the policy’s impact: whether the draft ever reaches the 35 nations, how Kazakhstan reacts when the pressure mounts, and the full composition of the signatory list. Until those details emerge, the United States’ attempt to force a geopolitical alignment in AI remains a high‑stakes gamble.

Questo articolo è stato scritto con l'assistenza dell'IA.
News Factory APP - notizie agentiche per potenziare il tuo SEO e AEO.