Sam Altman Reveals China's Rise in Open-Source AI Was Key Motivation Behind OpenAI's New Open-Weight Models

Sam Altman, CEO of OpenAI, revealed that a major factor behind the company’s decision to release its first open-weight models in years was the growing influence of Chinese open-source AI models. Speaking during a media briefing reported by CNBC, Altman said, “It was clear that if we didn’t do it, the world was gonna be mostly built on Chinese open-source models.” He added that while it wasn’t the only reason, it was a significant and pressing concern. OpenAI launched two new open-weight models on August 5: gpt-oss-120b, designed for high-end computing systems, and gpt-oss-20b, optimized to run on most desktops and laptops. This marked the first time since GPT-2 in 2019 that OpenAI has released a model with publicly available weights. Unlike closed models such as GPT-3, GPT-4, and the upcoming GPT-5, open-weight models allow developers to download, run, and fine-tune the AI locally, promoting transparency and innovation. Altman emphasized that the move aligns with OpenAI’s mission to ensure artificial general intelligence (AGI) benefits all of humanity. In an X post on the day of the launch, he stated that gpt-oss is “the best and most usable open model in the world.” He also highlighted the importance of building an open AI stack rooted in U.S. values and democratic principles, available freely to everyone. The timing of the release follows growing global competition in AI, particularly from China. In January, the Chinese startup DeepSeek unveiled R1, a powerful, open-weight model that gained widespread attention for its performance and affordability. The model’s success prompted U.S. political figures, including former President Donald Trump, to view it as both a positive development and a wake-up call for American tech leadership. During a Reddit AMA in January, Altman acknowledged that OpenAI had been slow to embrace open-source strategies. “I personally think we have been on the wrong side of history here and need to figure out a different open source strategy,” he said, signaling a shift in direction. While Meta’s Llama series has long been open-weight, CEO Mark Zuckerberg has recently stressed the need for caution in what is shared publicly. OpenAI’s latest move underscores a broader trend: as AI advances, the race to control foundational models is increasingly about access, values, and global influence.