Chinese open-source artificial intelligence technology is making significant inroads into the United States market, even as Washington and Beijing remain locked in a broader rivalry over advanced tech. Unlike the closed generative artificial intelligence models from OpenAI and Google, which keep their inner workings tightly controlled, Chinese providers such as Alibaba and DeepSeek are offering open models that allow programmers to customize parts of the software. China’s open-source approach is appealing to US developers and companies that do not always need the most cutting-edge capabilities but want tools that are flexible, inexpensive, and easier to adapt for specific applications.
The uptake has accelerated quickly. Globally, use of Chinese-developed open models has surged from just 1.2 percent in late 2024 to nearly 30 percent in August, according to a report cited in the article. One American entrepreneur said their business saves $400,000 annually by using Alibaba’s Qwen artificial intelligence models instead of proprietary systems, while still turning to OpenAI, Anthropic or Google when they need top-tier performance. US chip giant Nvidia, artificial intelligence startup Perplexity and California’s Stanford University are also using Qwen models in some of their work, underscoring how Chinese tools are being integrated into high-profile research and commercial environments.
The January launch of DeepSeek’s high-performance, low-cost and open source “R1” large language model challenged the perception that the best artificial intelligence technology must come from US companies and highlighted how far China has advanced. Other Chinese players, including MiniMax, Z.ai and Moonshot artificial intelligence with its agent-focused Kimi K2 model released in November, are pushing into areas such as autonomous agents that perform online tasks. While the Trump administration’s “AI Action Plan” in July called for “leading open models founded on American values,” major US firms are retreating from open-source, with Meta pivoting to closed models and OpenAI offering only limited “open-weight” releases. By contrast, the Chinese government has encouraged open-source artificial intelligence despite questions over immediate profits, and experts quoted in the article argue that transparency in open models improves scrutiny and data security, helping to build trust even as some Western clients remain wary of geopolitical and sanctions risks.
