NextFin News - On February 10, 2026, U.S. Commerce Secretary Howard Lutnick confirmed during a Senate hearing that the administration of U.S. President Trump will not block American firms from purchasing Nvidia’s advanced artificial intelligence processors, even as it maintains rigorous guardrails on exports to China. This announcement follows a pivotal January 13, 2026, policy shift by the Bureau of Industry and Security, which moved from a "presumption of denial" to a "case-by-case review" for high-end semiconductors. Under these new terms, Nvidia is permitted to sell its H200 chips to Chinese entities, provided they adhere to a 50% volume cap and a 25% tariff. However, the company currently faces delays in fulfilling orders for major Chinese clients, including ByteDance and Alibaba, due to unresolved disputes over "Know-Your-Customer" (KYC) requirements designed to prevent the diversion of technology to the Chinese military.
The geopolitical tightrope comes at a time when Nvidia’s market capitalization has stabilized between $4.3 trillion and $4.6 trillion, cementing its status as the world’s most valuable enterprise. According to FinancialContent, Nvidia’s revenue for the third quarter of fiscal 2026 reached a record $57.0 billion, driven by the global transition toward "Agentic AI" and the rollout of its Blackwell architecture. Despite this financial strength, the company’s stock dipped 0.7% to $188.68 following Lutnick’s remarks, reflecting investor anxiety over the sustainability of Chinese revenue streams. China has historically accounted for approximately 20-25% of Nvidia’s data center revenue, but that share is under pressure from both U.S. regulatory friction and Beijing’s own push for technological self-reliance.
The Trump administration’s strategy appears to be one of "transactional containment." By allowing the sale of the H200—a chip significantly more powerful than domestic Chinese alternatives but trailing Nvidia’s latest "Rubin" architecture—the U.S. aims to monetize Chinese demand while keeping a generational lead in compute capacity. According to InvestingLive, Chinese tech giants have already signaled intent to purchase over two million H200 chips in 2026, a windfall potentially worth $14 billion. However, this opening is far from a return to free trade. The mandatory third-party U.S. testing phase and the threat of the "AI Overwatch Act"—a bill passed by the House Foreign Affairs Committee on January 22, 2026—introduce a layer of legislative risk that could revoke export licenses with 30 days' notice.
From an analytical perspective, Nvidia is caught in a pincer movement between U.S. national security mandates and China’s accelerating indigenous innovation. While Nvidia currently holds an estimated 90% share of the AI accelerator market, the "moat" provided by its CUDA software ecosystem is being tested. In December 2025, reports surfaced that Chinese scientists in Shenzhen successfully developed a prototype Extreme Ultraviolet (EUV) lithography machine. While mass production is not expected until 2028, this breakthrough signals that the window for U.S. "chokepoint" diplomacy may be narrowing. If China successfully bypasses Western lithography constraints, Nvidia’s specialized "China-clear" chips could lose their competitive edge against domestic processors that do not carry the burden of U.S. tariffs or surveillance requirements.
Furthermore, the macro-economic backdrop adds a layer of complexity to Nvidia’s valuation. As of February 2026, China has reduced its holdings of U.S. Treasuries to a 17-year low of $682 billion. This "stealth selling," combined with Beijing’s recent directives advising domestic firms to avoid U.S. chips unless "absolutely necessary," suggests a strategic decoupling that transcends mere trade policy. For Nvidia, the risk is no longer just about what it is allowed to sell, but whether its largest customers in the East are being incentivized to build an entirely parallel AI ecosystem. Analysts at Janney Montgomery Scott note that while the "Inference" boom—the phase where AI models are deployed at scale—is driving short-term demand, the long-term trajectory depends on whether Nvidia can transition from a hardware provider to a global AI operating system.
Looking forward, the H2 2026 launch of the Rubin architecture will be the ultimate litmus test for Nvidia’s trillion-dollar valuation. If the U.S. President continues to use chip exports as a bargaining chip for broader trade concessions, Nvidia may find itself perpetually navigating a fragmented global market. The company’s ability to maintain its 73.4% gross margins will depend on whether it can keep the "Sovereignty AI" movement—where nations build domestic clouds—tethered to its proprietary hardware. In the immediate term, all eyes remain on Nvidia’s February 25 earnings call, where the company is expected to provide clarity on how the new Trump-era guardrails will impact its fiscal 2027 outlook and its precarious balance of power in the Pacific.
Explore more exclusive insights at nextfin.ai.
