Sam Altman, the CEO of OpenAI, has openly voiced his growing concerns about China’s progress in artificial intelligence, warning that the United States may be underestimating its rival’s next-generation AI capabilities.
During a recent media briefing, Altman emphasized that the US-China AI race is far more complex than a simple comparison of “who is ahead.” Instead, it spans multiple layers—including inference capacity, research innovation, and product development.
“I’m worried about China,” Altman admitted. “There’s inference capacity, where China probably can build faster. There’s research, there’s product; a lot of layers to the whole thing. I don’t think it’ll be as simple as: Is the U.S. or China ahead?”
China’s Open-Source AI Push Shapes OpenAI’s Strategy
One of the most striking revelations from Altman was that China’s rapid progress with open-source models like DeepSeek and Kimi K2 influenced OpenAI’s own decision to release its open-weight models—gpt-oss-120b and gpt-oss-20b.
Altman explained that without this move, global AI development might lean heavily toward Chinese open-source systems.
“These models were released because it was clear that if we didn’t do it, the world was going to be mostly built on Chinese open-source models,” he said, highlighting the strategic importance of staying competitive in this fast-evolving space.
The newly launched text-only models are designed with accessibility in mind:
- gpt-oss-120b: With 117 billion parameters, it can run on a single 80GB GPU, delivering benchmark results on par with or better than OpenAI’s o4-mini model.
- gpt-oss-20b: Featuring 21 billion parameters, it requires as little as 16GB of RAM, making it an affordable option for developers working with limited hardware resources.
These cost-effective models are set to empower researchers, startups, and enterprises worldwide, ensuring OpenAI maintains a significant role in shaping the global AI landscape.
Export Controls and the Semiconductor Challenge
Altman also raised doubts about the effectiveness of current US export restrictions on semiconductors, designed to limit China’s access to advanced AI chips. He suggested that such measures may only push China to accelerate its domestic semiconductor capabilities.
“My instinct is that it doesn’t work,” he noted. “You can export-control one thing, but maybe not the right thing… Maybe people build fabs or find other workarounds. I’d love an easy solution. But my instinct is: That’s hard.”
This observation comes at a time when Chinese tech giants like Huawei are aggressively investing in self-reliance. Huawei’s latest high-performance AI chip, the Ascend 910C, is reportedly designed to rival Nvidia’s H100—a move that could significantly reduce China’s dependence on US technology while positioning its chip industry as a global competitor.
The Bigger Picture: Risks for the United States
Industry analysts caution that tightening restrictions on China could backfire against US companies, limiting their access to one of the world’s most lucrative technology markets while fueling innovation in China’s semiconductor and AI sectors.
As the AI race intensifies, Altman’s warnings underscore a critical point: The future of global AI dominance may not be decided by a single breakthrough or company, but by how nations balance innovation, openness, and strategic independence.
For now, one thing is clear—China’s determination to build next-generation AI and semiconductor technology is reshaping the global tech landscape, and the United States will need to respond with more than just regulations if it hopes to stay ahead.
Read more :https://globalbizoutlook.com/junghee-kim-appointed-head-of-marketing-for-south-korea-at-chagee/
