Nvidia CEO Huang Uncertain About China’s Reception of H200 Chips Following Trump Meeting

Reading Time: 3 minutes

Nvidia Chief Executive Officer Jensen Huang expressed uncertainty regarding whether China would accept the company’s H200 artificial intelligence chips, following a private meeting with U.S. President Donald Trump on Wednesday.

Speaking briefly to reporters at the U.S. Capitol, Huang confirmed that export controls were discussed during the meeting but did not disclose further details. His comments come as the Trump administration weighs the possibility of allowing the sale of H200 chips to China.

“We don’t know. We have no clue,” Huang told reporters, adding that Nvidia cannot offer downgraded versions of the chips as Chinese buyers are unlikely to accept them. His remarks came just before a closed-door session with the Senate Banking Committee, which oversees export control policy.

The White House declined to comment on the private meeting.

If approved, the sale of H200 chips to China would represent a significant victory for Nvidia, the world’s most valuable semiconductor company. Nvidia has been advocating for eased restrictions, arguing that tighter controls benefit Chinese competitors such as Huawei Technologies.

Huang, who has reportedly built a strong working relationship with Trump since the 2024 election, emphasized that he visits Washington at the president’s request. He also met informally with lawmakers during his visit to the Capitol.

His trip coincided with a legislative decision in Congress to exclude a proposal known as the Gain AI Act from key defense legislation. That measure would have prioritized U.S. customers for advanced chips and restricted sales to China and other nations under arms embargoes.

Senator Mike Rounds, a Republican member of the Banking Committee, acknowledged Nvidia’s push for global competitiveness, noting that both Huang and lawmakers share concerns about export control implications. “They want the customers around the world. We understand that,” Rounds said. “At the same time, we’re all concerned, including Jensen, about restrictions regarding China.”

Senator Cynthia Lummis described the committee’s meeting with Huang as “educational” and confirmed that the Gain AI Act was not discussed during the session.

Any relaxation of current export rules would represent a departure from measures introduced in 2022 aimed at limiting Beijing’s access to the most powerful U.S. technologies. Such a move is expected to face resistance from national security advocates in Washington.

Earlier this year, Nvidia received approval to sell its H20 chip in China. However, Beijing reportedly urged domestic companies to avoid the product in favor of chips made by local firms. More recently, Nvidia failed to secure U.S. authorization to export a limited version of its latest Blackwell-generation chip during a meeting between President Trump and Chinese President Xi Jinping in October.

The H200 chip, which began shipping in 2024, is designed to support both the training and deployment of advanced AI models. Lawmakers across party lines have cited the potential sale of this processor as a key reason to revive the Gain AI Act.

Senator Elizabeth Warren, ranking Democrat on the Banking Committee, issued a letter to Commerce Secretary Howard Lutnick on Wednesday, urging the administration to maintain current limits on chip sales to China. She raised concerns about a lack of transparency in export control decisions and criticized efforts by large technology companies to expand into adversary markets.

“We should not allow Big Tech firms like Nvidia to sell sensitive technology to governments that do not share our values,” Warren wrote. The letter was co-signed by Representative Andy Kim.

In a recent interview, Huang acknowledged that China could represent a $50 billion market for Nvidia. However, the company has not included projected data center revenue from China in its current financial outlook.

“We would love the opportunity to re-engage the Chinese market,” Huang said, adding that the global use of open-source Chinese models demonstrates the broader impact of such technologies.