這篇文章探討了人工智慧在政治、戰爭與治國理政上日益真實且深遠的影響。隨著美國成功利用人工智慧在委內瑞拉與伊朗進行軍事行動，以及人工智慧模型展現出超人類的網路能力，全球力量平衡正處於關鍵時刻。雖然美國目前在軍事應用與前沿模型上保持領先優勢，但其主導地位正面臨中國快速追趕的嚴峻挑戰。

美國正遭遇多重戰略困境，其中最主要的是「擴散難題」，即中國透過提供成本較低的開源模型，成功將其影響力滲透至全球南方及已開發國家企業。同時，美國旨在限制中國發展的出口管制措施開始鬆動，國內對於人工智慧安全的擔憂引發政治分歧，且民主盟友間的團結也因對美國政策的疑慮而出現裂痕。

為了在未來的人工智慧競賽中保持優勢，美國必須重新調整其戰略。這包括強化出口管制以保護運算優勢，建立快速通道以協助盟友取得人工智慧基礎設施，並在國內制定透明的安全監管政策。作者強調，美國能否維持全球技術霸權，取決於政府是否能克服內部政策混亂，並有效應對這些日益複雜的挑戰。

This article explores the increasingly real and profound impact of artificial intelligence on politics, warfare, and statecraft. As the United States successfully utilizes AI for military operations in Venezuela and Iran, and as AI models demonstrate superhuman cyber capabilities, the global balance of power has reached a critical juncture. Although America currently maintains a leading edge in military applications and frontier models, its dominance is facing severe challenges from China's rapid advancements.

The United States is encountering multiple strategic dilemmas, most notably the "diffusion dilemma," where China successfully permeates its influence into the Global South and developed enterprises by offering cost-effective open-weight models. Simultaneously, US export controls aimed at restricting China's development are loosening, domestic anxieties regarding AI safety are sparking political divisions, and democratic allied solidarity is fracturing due to concerns over American policies.

To maintain its supremacy in the future AI race, the United States must recalibrate its strategy. This includes strengthening export controls to protect its computing advantage, establishing fast-track procedures to help allies acquire AI infrastructure, and enacting transparent safety regulations domestically. The author emphasizes that whether America can sustain its global technological hegemony depends on the government's ability to overcome internal policy chaos and effectively address these increasingly complex challenges.