Nvidia CEO Jensen Huang Joins Trump's China Delegation After Last-Minute Invitation

On the morning of May 13, 2026, Jensen Huang boarded a plane to Anchorage, Alaska — not to visit the state, but to intercept Air Force One. Roughly 24 hours earlier, the Nvidia CEO had been conspicuously absent from the White House's official roster of business executives accompanying President Donald Trump on his state visit to Beijing [1][2]. Then Trump saw the headlines, picked up the phone, and told Huang he wanted him along [3].

The reversal — from deliberate exclusion to personal presidential invitation in under a day — is more than a colorful diplomatic anecdote. It crystallizes the contradictions at the heart of U.S. China policy on artificial intelligence: an administration that has restricted Nvidia's most lucrative export market while simultaneously needing the company's CEO in the room to project commercial strength.

The Snub and the Callback

The initial decision to leave Huang off the delegation list was not an oversight. According to Semafor, Huang had previously discussed with Trump how his presence in Beijing could invite "awkward conversations" about the contested sale of Nvidia chips to China [4]. Trump's decision in late 2025 to allow limited chip sales to China under a fee-based model had already drawn sharp criticism from Republican China hawks, and bringing Huang to Beijing risked making semiconductor policy the centerpiece of media coverage rather than the broader trade agenda [4][5].

But the media narrative that formed around Huang's absence — framed as a "snub" by Bloomberg, CNBC, and Tom's Hardware — apparently bothered Trump more than the awkwardness he had been trying to avoid [1][6]. Trump called Huang directly, and by Tuesday evening, the CEO was en route to Alaska to join Air Force One for the second leg of the flight to Beijing [3]. Trump later told reporters he was "honored" to be accompanied by "the great Jensen Huang" [7].

Who Was in the Room

The delegation that gathered around Trump for the May 13–15 visit to Beijing reads like a Fortune 500 board meeting. Among the roughly 16 executives: Elon Musk of Tesla and SpaceX, Tim Cook of Apple, Larry Fink of BlackRock, David Solomon of Goldman Sachs, Jane Fraser of Citigroup, Stephen Schwarzman of Blackstone, Cristiano Amon of Qualcomm, Sanjay Mehrotra of Micron, Kelly Ortberg of Boeing, and Brian Sikes of Cargill [8][9].

The composition is revealing. Nearly every executive present leads a company with substantial China exposure. Tesla operates a gigafactory in Shanghai that produces roughly half the company's global vehicles. Apple assembles the majority of its products in China. BlackRock and Goldman Sachs have expanded their Chinese asset management and investment banking operations. Qualcomm and Micron, like Nvidia, operate under the shadow of semiconductor export controls [8][10].

The pattern suggests the delegation was curated not for symbolic diversity but for commercial leverage — executives whose firms have the most at stake in the U.S.-China economic relationship, and therefore the most incentive to deliver goodwill. Critics might call this a lobbying trip with diplomatic cover. Defenders would argue it is standard practice for trade missions dating back decades.

The $17 Billion Question: Nvidia's China Dependency

To understand why Huang's presence mattered, follow the money.

Source: Nvidia SEC Filings / MacroTrends

Data as of Feb 1, 2026CSV

Nvidia's total revenue has grown at a staggering pace — from $26.9 billion in fiscal year 2022 to $215.9 billion in fiscal year 2026, driven overwhelmingly by AI data center demand [11]. But the China portion of that revenue tells a different story: one of regulatory whiplash and collapsing market access.

Source: Nvidia SEC Filings / Analyst Estimates

Data as of Feb 1, 2026CSV

In fiscal year 2025 (ending January 2025), Nvidia generated an estimated $17.1 billion from China-related customers, representing roughly 13% of total revenue [12]. By fiscal year 2026, after the H20 export ban took full effect, that figure cratered to an estimated $7.5 billion [12]. Nvidia CEO Huang himself told analysts in mid-2025 that China revenue had "effectively dropped to zero" for new chip shipments, though some legacy product and licensing revenue continued [13].

The financial damage was concrete. In April 2025, after the U.S. government imposed licensing requirements on the H20 — a chip Nvidia had specifically designed to comply with earlier export restrictions — the company took a $5.5 billion charge related to excess inventory and purchase commitments [14]. The H20 had generated an estimated $12 billion to $15 billion in revenue in 2024 alone [14]. Nvidia's SEC filings explicitly identify China export controls as a material risk, noting an "$8.0 billion loss in H20 revenue" from the latest round of restrictions [15].

Huang's description of China as a "$50 billion opportunity" in August 2025 [16] was not hyperbole — it was a measure of what Nvidia stood to gain if restrictions eased, and what it was leaving on the table every quarter they didn't.

Export Controls: The Policy Landscape

The export control regime governing Nvidia's China sales has evolved through several phases. The Biden administration's October 2022 rules first restricted advanced AI chips above certain performance thresholds. Nvidia responded by designing the H800 and then the H20 — chips deliberately hobbled to fall below the thresholds. The H20 became enormously popular among Chinese AI firms, including DeepSeek, whose competitive R1 model was trained partly on H20 hardware [14][17].

In April 2025, the government closed this workaround by requiring export licenses for H20 shipments to China [14][15]. The move effectively shut down Nvidia's remaining compliant product line for the Chinese market. By August 2025, limited licenses were granted allowing roughly $60 million in sales to select customers — a rounding error against the billions in lost revenue [15]. U.S. officials also signaled an expectation that the government would receive 15% or more of revenue from any licensed sales, a fee-based model that Trump later expanded [15][18].

In late 2025, Trump authorized Nvidia to ship H200 chips to "approved customers" in China, with a 25% government cut [18]. Whether the Beijing trip would produce further relaxation of these controls was the central question hanging over Huang's participation.

Did Huang Get Anything?

As of the delegation's arrival in Beijing, no specific licensing approvals, chip-sale agreements, or policy rollbacks tied to Nvidia had been announced [7][18]. Trump himself acknowledged that "Nvidia and its Blackwell chips didn't come up" between him and Xi Jinping [18].

The stock market, however, read the tea leaves optimistically. Nvidia shares rose approximately 1.5% in overnight trading after Trump confirmed Huang was on the trip, and gained over 2% during the following trading session [7][19]. The market reaction reflected not any concrete policy change but the symbolic signal that Nvidia was at the table — and therefore unlikely to be further punished by the administration's China policy.

Analysts at TipRanks noted that Nvidia stock had been underperforming the SOXX semiconductor index by a historically wide margin, and that Huang's inclusion in the delegation represented a potential catalyst for narrowing that gap [19].

The Huawei Factor

Every day that Nvidia is locked out of the Chinese market is a day that Huawei gains ground. The Chinese telecom giant has been aggressively expanding its Ascend AI chip line, with revenue projected to hit $12 billion in 2026 — up from $7.5 billion in 2025 and $4.5 billion in 2024 [20][21].

Source: Bloomberg / Analyst Estimates

Data as of Apr 1, 2026CSV

Huawei plans to deliver roughly 600,000 units of its flagship Ascend 910C in 2026, double the 2025 level, and has the Ascend 950PR slated for release in early 2026 [20][22]. IDC data shows Nvidia's China AI chip market share fell from 85% to 70% in 2024, while domestic Chinese chip brands collectively reached 30% share [21]. The longer U.S. export restrictions persist, the more permanent that market share shift becomes.

This dynamic creates a genuine national interest argument for Huang's presence in Beijing. If the policy goal is to prevent China from developing an entirely self-sufficient AI chip ecosystem, then keeping some level of U.S. commercial engagement — even restricted — maintains American firms' foothold and slows Huawei's displacement of Western technology. Former Commerce Department officials have made this case publicly: total decoupling accelerates, rather than retards, China's technology independence [17].

The Case For and Against CEO Diplomacy

The skeptic's case is straightforward: a CEO whose company stands to gain billions from relaxed export controls should not be sitting in on diplomatic negotiations where those controls are discussed. The appearance of private commercial interests being advanced through state-level access is a recurring concern in trade policy. There is no formal legal prohibition on it — the Federal Advisory Committee Act and conflict-of-interest statutes apply to government employees, not private executives invited on a trip — but ethics scholars have long argued that the practice creates at minimum an optics problem and at worst a policy-capture risk [23].

The defender's case is equally coherent: American executives provide real-time commercial intelligence to U.S. negotiators who might otherwise lack granular understanding of on-the-ground market conditions. Tim Cook's successful lobbying during Trump's first term — which secured tariff exemptions for ten Apple product categories by 2019, after Cook personally argued that tariffs would advantage Samsung — is cited as a case where CEO access produced outcomes aligned with broader U.S. economic interests [24]. Cook's presence on this delegation, alongside Huang, suggests the pattern is intentional.

Historical precedent supports both readings. During the 2018–2019 U.S.-China trade war, multiple CEO roundtables were convened, and executives from Boeing, Qualcomm, and agricultural firms joined trade missions. In those cases, the concrete outcomes were mixed: China made large agricultural purchase commitments (which it partly fulfilled) while structural issues like intellectual property theft and forced technology transfer remained largely unresolved [25].

How Beijing Reads the Room

Chinese state media coverage of Huang's prior April 2025 visit to Beijing was notably warm. Vice Premier He Lifeng personally told Huang that China "welcomed more U.S. companies including Nvidia to further explore the Chinese market" at the Great Hall of the People, and state broadcaster CCTV quoted Huang saying "We hope to continue to cooperate with China" [26]. The messaging framed Huang's visit as evidence of American commercial dependency on Chinese demand — a narrative that serves Beijing's negotiating position.

Huang's presence on the Trump delegation likely reinforces that framing. From Beijing's perspective, the inclusion of executives from companies most affected by trade restrictions signals that Washington is feeling economic pressure from its own policies. Whether that reading is accurate or not, it shapes the negotiating dynamic: China can point to the delegation roster as evidence that decoupling is unsustainable, and use that leverage to resist concessions on issues like industrial subsidies, intellectual property, and military-civil fusion — the very concerns that motivated export controls in the first place [27].

The Broader Trade Context

The delegation arrives in Beijing at a moment of cautious détente. After tariffs on both sides peaked at 125% in April 2025, the U.S. and China agreed to a 90-day reduction to 10% in May 2025, extended it in August, and then formalized a one-year extension in November 2025 that runs through November 2026 [25]. Fentanyl-related tariffs were reduced from 20% to 10%, and China committed to purchasing at least 25 million metric tons of U.S. soybeans annually through 2028 [25].

But major friction points persist. Semiconductor export controls remain in place. Rare earth mineral access — critical for U.S. defense and technology supply chains — is a source of ongoing tension. And the fundamental question of AI governance, including whether China's military-linked entities should have any access to advanced computing hardware, remains unresolved [25][27].

Trump's approach to the summit has shifted from the confrontational posture of early 2025 to one that the Washington Post described as "focused on getting along" [28]. Whether that recalibration reflects strategic calculation, commercial pressure from the executives now flanking him, or both, is a question that will define not just the outcome of this trip but the trajectory of U.S.-China economic relations through the 2026 midterm elections.

What Comes Next

The fact that Huang was initially excluded, then personally summoned by the president, tells us something about the unstable equilibrium of current U.S. semiconductor policy. The administration wants to restrict China's access to advanced AI chips for national security reasons but cannot ignore the commercial consequences for its most valuable technology company. Nvidia wants access to a market that could be worth $50 billion but cannot afford to be seen as undermining national security. And China wants to buy American chips while simultaneously racing to make them irrelevant.

Huang's seat on Air Force One does not resolve any of these tensions. But it ensures that when the next round of export control decisions is made, the CEO of the world's most important AI hardware company will have been in the room — and that both Washington and Beijing know it.