Nvidia launched a new AI chip platform to maintain leadership as competition intensifies.

AI juggernaut Nvidia unveiled its latest AI platform on Monday at the annual Consumer Electronics Show in Las Vegas, where CEO Jensen Huang's keynote was an early must-see event at the globe's biggest tech showcase. With its new Vera Rubin product, first announced in September, Nvidia is seeking to lock in its dominance of the AI chip business. The company currently holds an estimated 80% of the global market for AI data center chips.

Named after American astronomer Vera Rubin, the new model marks a profound shift from Nvidia's previous generation of AI architecture, Blackwell, which launched in late 2024. The Rubin-based products will be available from partners in the second half of 2026. Nvidia promises that the Rubin product will run five times more efficiently than previous offerings, a key performance metric as the energy needs related to AI become an increasing concern. The platform comprises "six chips that make one AI supercomputer," according to Dion Harris, Nvidia's director of data center and high-performance computing. These will "meet the needs of the most advanced models and drive down the cost of intelligence," Harris added.

However, Nvidia faces mounting pressure from multiple fronts. Traditional chip-making rivals like AMD and Intel are pushing hard to compete. More significantly, Nvidia's biggest customers—Google, Amazon, and Microsoft—are increasingly developing their own chips to reduce their dependence on the company. Google's latest AI model, Gemini 3, was trained without Nvidia's technology. Meanwhile, China is racing to build domestic alternatives to Nvidia products, which have faced US export restrictions, hobbling the Chinese tech sector.

Since ChatGPT's release in 2022, Nvidia has been updating its product line at a furious pace, raising questions about whether tech companies building AI models can afford to keep their technology state-of-the-art.