Analyzing the latest market signals, I conclude that Nvidia is taking a strategically important step by signing a $6 billion agreement with the AI startup Poolside. The goal is to develop a powerful open-weights model that can directly compete with Chinese large language models (LLMs) such as DeepSeek and Kimi K3. This is not just a deal, but a clear signal of shifting priorities in the race for dominance in the field of artificial intelligence.
As early as August 21, information emerged that the chipmaker would pay Poolside $6 billion for licenses to AI models. However, the terms do not end there: Nvidia will also offer jobs to more than 100 of the startup's employees and additionally invest $1 billion. It is obvious that the company is betting on open source, seeking to democratize access to advanced technologies and prevent their concentration in the hands of a narrow circle of giants. This is also a direct way to stimulate future demand for its own chips and AI systems, creating an ecosystem dependent on its hardware.
The structure of the Poolside deal resembles Nvidia's recent acquisition of chip developer Groq, where $20 billion was paid for the startup's licenses, followed by the hiring of most employees and founders. This approach allows Nvidia to quickly integrate talent and technology, bypassing the lengthy processes of corporate mergers.
In my assessment, this agreement is a direct challenge not only to Chinese developers, but also to American leaders like OpenAI and Anthropic. Open-weights models are significantly cheaper to operate and allow for easy customization, making them extremely attractive to businesses. This could undermine the commercial model of closed development that currently dominates the market.
The context of the deal is also important. In July, discussions resumed within the administration of U.S. President Donald Trump about possible restrictions on the use of Chinese open-weight models. Nvidia, Meta, and Microsoft, along with 22 other companies, opposed such measures, warning that bans would not strengthen U.S. leadership but would merely shift the market toward a few closed developers. The escalation of the discussion was largely provoked by the release of Kimi K3, whose creators the White House accused of distilling Anthropic's Fable model.
My conclusion: Nvidia's investment in open models is a forward-looking move that could reshape the AI landscape. However, success will depend on whether the company can not only create a competitive model, but also build a sustainable ecosystem around it capable of withstanding both Chinese and American closed giants.