The artificial intelligence market continues to be in turmoil. Nvidia, the largest manufacturer of AI chips, is betting on open models and preparing to directly challenge Chinese developments such as DeepSeek and Kimi K3. This involves a deal with the AI startup Poolside, with the amount reaching $6 billion. This is not just a licensing agreement—it is a strategic maneuver that could reshape the balance of power in the industry.
The terms of the partnership are impressive: Nvidia will pay Poolside $6 billion for licenses to their AI models, offer jobs to more than 100 startup employees, and additionally invest $1 billion in development. This deal structure resembles the recent acquisition of chip developer Groq, where Nvidia paid $20 billion for licenses and integrated most of the specialists and founders into its workforce. It is clear that the company is systematically buying up talent and technology to strengthen its position in the AI arms race.
Open Source as a Weapon
The key goal of the deal is to create a powerful model with open weights that will compete with Chinese LLMs. This is a fundamental stance. Nvidia is betting on the democratization of technology: open models can be freely used, refined, and adapted for specific tasks. This approach not only lowers barriers for developers but also undermines the dominance of closed ecosystems controlled by a few giants like OpenAI and Anthropic.
Strategically, this is a brilliant move. By supporting open source, Nvidia stimulates demand for its chips and AI systems. The more models with open weights there are, the more companies and developers need computing power, and therefore Nvidia's products. This creates a closed loop that strengthens the manufacturer's business.
Political Context
The deal comes amid heightened debate in the U.S. administration about possible restrictions on the use of Chinese open-weight models. In July, discussions resumed about measures that could complicate access to developments from China. However, 25 American companies, including Nvidia, Meta, and Microsoft, opposed the bans, warning that such steps would not strengthen U.S. technological leadership but would merely shift the market toward a few closed developers. It was the release of Kimi K3 from Moonshot AI, which the White House accused of distilling Anthropic's Fable model, that became the catalyst for these debates.
My analysis shows: Nvidia is not just protecting its commercial interests but also trying to shape a new technological order where openness becomes the main competitive advantage. In the long term, this could lead to fragmentation of the AI market, but for now, the bet on open source looks like the most pragmatic and scalable path. The only question is whether American companies can create a truly competitive alternative to Chinese models, which often offer comparable quality at lower costs.