Jun 3, 2026 · 11:44 PM
Subscribe
Home Ai

Nvidia loses a seat at the table on Trump's China trip

Jensen Huang was not invited to President Trump's China business delegation even as Nvidia remains central to the AI chip fight. The omission highlights a bigger risk for founders: policy access, export controls and customer concentration can shape growth as much as product demand.

Julian Lim
· 5 min read · 479 views
Nvidia loses a seat at the table on Trump's China trip

Jensen Huang has been left out of President Trump's China business delegation at a moment when Nvidia needs political access almost as much as customer demand.

Nvidia is the company most closely identified with the AI chip race, but its chief executive is not expected to be in Beijing when President Donald Trump arrives this week for talks with China. That is the interesting part. The trip is being built around deals, exports and commercial diplomacy, yet the company at the center of the most sensitive technology fight in the world appears to be watching from the outside.

According to a report from Reuters, Jensen Huang was not invited to join the China visit, with the White House placing more emphasis on agriculture and commercial aviation, including possible Boeing aircraft orders. Other business leaders were expected to be part of the wider delegation, including Apple CEO Tim Cook and Tesla CEO Elon Musk, while separate reports also pointed to senior figures from finance, aviation and chip-adjacent industries.

That does not mean Nvidia has suddenly lost Washington. Huang has built a strong relationship with Trump, and Nvidia remains one of the most important companies in the U.S. economy. But access is never permanent. For founders, this is the useful lesson. A company can be strategically vital, politically visible and commercially dominant, and still find itself absent from the room when the agenda changes.

The timing matters because Nvidia's China story is already complicated. Trump previously gave the green light for Nvidia's H200 AI chips to be exported to approved Chinese customers, with conditions attached. That decision was a major reversal from the tighter controls that had limited China's access to advanced U.S. AI processors, and it immediately raised concerns among China hawks who fear that powerful chips could strengthen Beijing's military and surveillance capabilities.

Yet approval on paper has not turned into sales. Commerce Secretary Howard Lutnick told lawmakers on April 22 that Nvidia had not yet sold H200 chips to Chinese companies, saying Chinese buyers still faced permission hurdles from their own government. His explanation was simple enough: Beijing wants domestic companies to keep spending on local chip suppliers instead of relying on Nvidia.

That leaves Nvidia in an awkward position. The U.S. government has opened a narrow door, but China may not want its companies to walk through it. The result is a market that looks valuable in theory and unreliable in practice. For a business built on scale, product cycles and long-term customer commitments, that is not a small problem.

China was once too important for Nvidia to ignore, and it still represents one of the largest pools of AI demand in the world. But the market is no longer just a sales opportunity. It is a policy test, a national security argument and a bargaining chip in a broader trade relationship. That changes how investors should read every headline about permissions, licenses and delegations.

Policy access is now a business risk

Founders often think about regulation as something that happens after a company grows large enough to attract attention. Nvidia shows why that thinking is too narrow. In sectors such as AI chips, defense software, energy, fintech and biotech, policy is not an afterthought. It can decide who gets to sell, where they can sell and how quickly revenue can turn into cash.

Huang's absence from the China trip may not damage Nvidia's near-term business. The company still has enormous demand from cloud providers, enterprise customers and governments racing to build AI infrastructure. Its most advanced Blackwell systems remain the kind of product that customers plan around, not a commodity that can be easily swapped out.

But diplomatic access has a different value. It lets a company shape how policymakers understand trade-offs. It gives executives a chance to explain why a product matters, what restrictions cost and how competitors might benefit from overly blunt rules. When a company is not present, others get more room to define the agenda.

That is especially relevant when the White House is emphasizing aircraft, soybeans and broad commercial wins. Those are tangible trade items. They fit neatly into negotiations. AI chips are more difficult because they sit between commerce and security. Selling them can help an American company, support U.S. technological influence and generate government revenue under certain arrangements. Selling them can also trigger warnings about military use and strategic leakage.

For Nvidia, the risk is not that one missed trip changes everything. The risk is that China becomes a market where every sale requires two governments to stay aligned long enough for customers to act. That is a fragile foundation for growth. It also gives local Chinese chipmakers more time to close the gap, even if they still trail Nvidia at the high end.

There is a broader lesson here for any company selling into politically sensitive markets. Customer concentration is not only about having too much revenue tied to one buyer. It can also mean having too much of your future tied to one jurisdiction, one regulator or one diplomatic relationship. When that relationship shifts, the spreadsheet changes quickly.

The next thing to watch is whether the Trump trip produces any signal on AI chip exports or whether chips remain pushed behind easier trade wins. If the visit delivers progress for aviation and agriculture while Nvidia's China approvals remain stuck, the message will be clear enough. In the new AI economy, having the best product is powerful, but it does not guarantee a seat at every table.

Also read: China is turning AI demand into a new export engineAMP is turning AI compute into a new venture capital battlegroundClaude adds /goal to keep working until the job is done

TOPICS
Julian Lim is an entrepreneur, technology writer, and a researcher. He started JL Data Analysis after graduating from NUS in Intelligent Systems. Julian writes about technology innovations and entrepreneurship on Business Times, Asia Pacific Magazine and occasionally contributes to Startup Fortune.
Related Articles
More posts →
Loading next article…
You're all caught up