Nvidia Is Making a Massive $26 Billion Bet on the Future of Artificial Intelligence (AI)
Nvidia plans to invest $26 billion over the next five years to develop open-source AI models, enhancing its position in the AI market. This investment aims to create models that are free and accessible, allowing researchers and startups to build tailored AI systems. Nvidia's competitive edge lies in its CUDA programming platform, which optimizes GPU performance. The company's revenue has surged over 1,000% since early 2023, and its stock remains attractively priced at less than 38 times earnings, presenting a buying opportunity for investors.
For many investors, when the discussion turns to artificial intelligence (AI), Nvidia (NVDA +0.64%) is the first topic of conversation. The company pioneered the graphics processing units (GPUs) that revolutionized gaming with the introduction of parallel computing. This groundbreaking approach divided large computational tasks across multiple processors (or cores), dramatically reducing the time needed to complete the task. It turned out that parallel processing worked equally well on the large datasets necessary to facilitate AI.
Nvidia has ridden the AI revolution to new heights, helping it become the world's most valuable publicly traded company, with a market cap of $4.5 trillion.
Management is looking ahead, and Nvidia is planning a big bet on the future of AI.
Image source: Getty Images.
A $26 billion bet
Over the next five years, Nvidia plans to invest $26 billion to develop open-source AI models, according to a report that first appeared in Wired and later confirmed by Nvidia executives. By delving further into the software side of AI, Nvidia would be better positioned to develop, test, and scale next-generation AI models and systems aligned with its industry-leading AI chips.
The attraction of open-source models is simple: they are free and available to anyone who wants to use them, and give researchers, data scientists, and start-ups a jumping-off point to modify and build AI systems tailored to their own needs.
Many of the foremost models in the U.S. are proprietary and not accessible to the general public. For example, OpenAI's flagship GPT-5.4, Alphabet's Google Gemini 3.1 Pro, and Anthropic's Claude Opus 4.6 are premium models that generally require a paid subscription or cloud access to use.
The result is that many researchers and data scientists are building their AI systems on top of open-source models, and many of the freely available models were developed in China. If Nvidia were to develop its own open-source models, it could sync them to perform better with its hardware, providing a more unified experience.
Keen competitive advantage
Beyond the company's chip prowess, Nvidia has built a wide moat around its hardware through its Compute Unified Device Architecture (CUDA) programming platform and software architecture. CUDA is a suite of programming tools that helps developers optimize GPU performance. Nvidia boasts more than 400 libraries that help users "build, optimize, deploy, and scale applications across PCs, workstations, the cloud, and supercomputers using the CUDA platform."
These programming hacks help developers, data scientists, and other users harness the computational horsepower of the GPU, creating a durable competitive advantage as high switching costs keep users locked into its ecosystem.
If Nvidia offered its own open-source AI models tuned to its GPUs, this advantage would be increased by an order of magnitude.
NASDAQ: NVDA
Key Data Points
Paving the road to future success
Until now, Nvidia has relied primarily on its chipmaking expertise to power its blistering results, which have been nothing short of astonishing over the past few years. Since the dawn of the AI revolution in early 2023, its revenue has grown more than 1,000%, from $6 billion to $68 billion. At the same time, its net income has soared 2,940%, from $1.4 billion to $43 billion. This, in turn, has pushed its stock price up 1,170%.
However, as competition heats up and AI models become more fragmented, this strategy could help keep Nvidia at the cutting edge of AI development for years to come.
Despite its impressive run in recent years, Nvidia is still attractively priced. The stock is currently trading for less than 38 times earnings, a significant markdown compared to its five-year average multiple of 74. This gives astute investors the opportunity to pick up shares at a discount, even as Nvidia paves the road to future success.
