When we think about the global race for AI supremacy, the narrative often focuses on who has the best technology, the most funding, and the biggest names. While the United States is currently seen as the frontrunner, the situation is far more complex than it appears, according to none other than OpenAI CEO Sam Altman.
"I'm worried about China," Altman recently told CNBC. His concerns aren't about a simple technological gap that can be solved with a few export controls. Instead, he believes the race is multifaceted, with different countries excelling in different areas—from research to product development and, most importantly, inference capacity.
The Flaw in Export Controls
The debate over export controls on AI hardware has been front and center. The U.S. government banned the export of NVIDIA's powerful H20 AI GPUs to China in April 2025, a move intended to slow China's progress. However, this decision was quickly reversed after intense lobbying from U.S. tech giants like NVIDIA, who feared being locked out of the massive Chinese market.
Altman shares a similar view to NVIDIA CEO Jensen Huang, arguing that export controls are a flawed strategy. They can be easily circumvented through black markets or by driving foreign markets to develop their own domestic alternatives. The Financial Times reported that Chinese companies smuggled roughly $1 billion worth of NVIDIA chips in just the first three months following the ban. This shows that demand will find a way, even if it’s through unofficial channels.
Furthermore, Huang pointed out that blocking sales to China would be "a grave mistake," as it would empower Chinese firms like Huawei to develop their own, more widely adopted AI chips, potentially setting a new global standard.
China's Influence on U.S. Tech Strategy
The competitive pressure from China is so significant that it is directly influencing the decisions of U.S. companies. Take OpenAI's recent release of two open-source AI models, gpt-oss-120b and gpt-oss-20b. These were the first open-source releases from the company since 2019.
Why the change? Altman explained that a key factor was the realization that "the world was gonna head to be mostly built on Chinese open source models" if U.S. companies didn't step up. This strategic move wasn't just about innovation; it was a preemptive strike to prevent China from dominating the open-source AI ecosystem.
A Race with Many Lanes
Altman’s insights paint a picture of an AI race that isn’t a simple sprint but a multi-layered marathon. While the U.S. currently leads in certain aspects, China may be ahead in others, particularly in its ability to quickly scale up inference capacity.
The traditional tools of economic warfare, like export bans, may be ineffective in a domain as fluid and fast-moving as AI. Instead, they could inadvertently harm the very companies they are meant to protect, while pushing rivals to innovate faster. The real victory in this war might not be about who has the most advanced chip today, but who can adapt, innovate, and, most importantly, maintain a position of influence in a rapidly changing world.
CTA: I'm worried about China. What are your thoughts? Will export controls truly curb their AI progress, or will they only push China to find new, more powerful solutions? Let me know your thoughts in the comments below!