Nvidia is investing $6 billion in Poolside technology and talent to strengthen its Nemotron open-weight AI models and compete with China’s DeepSeek and Kimi.
Focus Keyphrase: Nvidia $6 billion AI investment
Secondary Keywords: Nvidia AI, Nvidia Nemotron, Poolside AI, open-weight AI models, U.S. China AI race, DeepSeek, Kimi K3, artificial intelligence, Jensen Huang, AI competition, American AI leadership
Categories: AI, Technology, Business, Artificial Intelligence, U.S.-China Competition
Tags: Nvidia, AI, Artificial Intelligence, Poolside, Nemotron, DeepSeek, Kimi K3, Open-Weight AI, Jensen Huang, China AI, U.S. Technology, AI Race, Semiconductors
Nvidia Makes a $6 Billion Move in the U.S.-China AI Race
Nvidia is making an aggressive new move in the global artificial intelligence race, committing $6 billion to a deal involving AI startup Poolside as the chip giant seeks to build one of the world’s most powerful open-weight AI models.
The strategy represents a significant expansion of Nvidia’s role in artificial intelligence. Rather than concentrating exclusively on the chips that power AI systems, Nvidia is increasingly investing in the software, models and developer ecosystem built around those processors.
According to The Wall Street Journal, Nvidia plans to license Poolside’s technology while bringing more than 100 Poolside employees, including engineers, into Nvidia. The engineers are expected to work on Nvidia’s Nemotron family of open-weight AI models.
The broader objective is clear: strengthen the American AI ecosystem while responding to the rapid progress of Chinese companies developing increasingly capable and relatively inexpensive open-weight models.
Nvidia and Poolside: What the $6 Billion Deal Means
The agreement involves significantly more than a conventional technology investment.
Nvidia plans to invest $1 billion in Poolside at a pre-money valuation of $12 billion, while separately paying approximately $6 billion to license Poolside’s technology and hire most of its engineering workforce, according to the Journal.
Poolside was founded in 2023 by Eiso Kant and Jason Warner, the former chief technology officer of GitHub. Its leadership is not expected to join Nvidia, while more than 100 employees will move to Nvidia to contribute to the company’s Nemotron project.
That distinction is important because Nvidia is effectively acquiring access to both advanced AI technology and specialized engineering talent without simply absorbing the entire startup.
For Nvidia, the transaction could accelerate development of increasingly sophisticated open-weight models at a time when AI competition is becoming more global.
Why Open-Weight AI Matters
The most important element of Nvidia’s strategy is its focus on open-weight AI.
Traditional proprietary AI systems are generally controlled by the companies that develop them. Users can interact with these models through applications or APIs, but they typically cannot freely download and modify the underlying model weights.
Open-weight models operate differently.
They can generally be downloaded, customized and adapted by developers, businesses, researchers and governments, depending on their licensing terms.
This can dramatically reduce barriers to AI deployment.
For businesses, open-weight models can offer greater control over infrastructure and data. For governments, they can potentially reduce dependence on a small number of foreign AI providers. For developers, they can create opportunities to modify models for specialized applications.
Nvidia CEO Jensen Huang has increasingly positioned open AI ecosystems as strategically important for American technological leadership. The company launched the Nemotron Coalition in March with organizations including Mistral, Thinking Machines Lab and Perplexity to encourage collaboration around open models.
China’s Open AI Models Are Driving the Strategy
Nvidia’s investment comes against the backdrop of intensifying technological competition between the United States and China.
The emergence of DeepSeek in early 2025 dramatically changed perceptions of China’s ability to develop advanced AI models efficiently.
DeepSeek’s low-cost, open-weight approach triggered major discussion across Silicon Valley and financial markets, demonstrating that Chinese AI developers could potentially produce highly competitive systems without simply replicating the spending strategies of the largest American AI laboratories.
Since then, additional Chinese AI developers have gained attention, including companies behind models such as Kimi and the GLM family.
The result is an increasingly competitive AI environment in which the United States is no longer competing only against large Chinese technology companies in computing hardware.
The competition increasingly involves:
- AI foundation models
- Open-weight models
- AI chips
- Data centers
- Computing infrastructure
- Developer ecosystems
- AI talent
- Energy and electricity
- Cloud computing
- Research and engineering capabilities
Nvidia’s Poolside transaction fits directly into this broader contest.
Nvidia Is Moving Beyond the AI Chip Business
Nvidia remains best known for designing GPUs, which became essential infrastructure for training and running modern AI systems.
But the company’s strategy is increasingly expanding beyond semiconductor hardware.
The company is developing CPUs and networking technologies, investing in AI startups, supporting data-center construction and building its own AI models.
That creates a much broader Nvidia ecosystem.
The Poolside agreement therefore matters because Nvidia is attempting to participate across multiple layers of the AI stack.
At the hardware level, Nvidia supplies processors.
At the infrastructure level, it supports data centers and computing capacity.
At the software level, it develops AI tools and model ecosystems.
And now, through Nemotron, it is attempting to become an increasingly important model developer.
Nemotron Could Become a Major American Open-Weight AI Platform
Nvidia introduced Nemotron as part of its effort to develop open-weight AI models.
The company has already released smaller versions, including Nemotron 3.5 Lightning, while reportedly developing a much larger model.
According to people familiar with the project cited by the Journal, the largest version under development could potentially contain more than one trillion training parameters.
Parameters are not a perfect measurement of AI capability, but the scale illustrates Nvidia’s ambition.
The company wants Nemotron to compete with sophisticated frontier models while providing a more open alternative.
The Poolside engineering team could help Nvidia accelerate this development.
That could make the next phase of the AI race less about individual chatbot brands and more about which ecosystems can provide powerful, customizable models to the largest number of developers and organizations.
The Risk: Nvidia Could Compete With Its Own Customers
Nvidia’s strategy also creates a complicated business dilemma.
Some of Nvidia’s biggest customers are AI companies such as OpenAI and Anthropic.
Those companies rely heavily on Nvidia hardware to train and operate their AI models.
Yet Nvidia is now developing AI models that could compete with proprietary systems produced by those same companies.
That creates an unusual relationship.
Nvidia is simultaneously:
- A supplier to leading AI companies.
- An investor in AI startups.
- A developer of its own AI models.
- A supporter of open-weight AI.
- A competitor in parts of the AI software market.
The situation could become even more complicated as major AI companies increasingly develop custom AI chips and alternative computing infrastructure.
In other words, Nvidia’s customers are becoming competitors while Nvidia itself is becoming more vertically integrated.
Poolside’s Computing Challenge
The Poolside deal also illustrates one of the biggest problems facing AI startups: access to computing power.
Poolside reportedly attempted to raise $2 billion in 2025 to finance a large Nvidia server cluster but did not complete the fundraising in time.
The company subsequently faced concerns about whether it would have sufficient computing capacity to continue scaling its AI models.
Nvidia, by contrast, has enormous access to computing resources, capital and AI infrastructure.
That makes the partnership strategically attractive to both sides.
Poolside contributes technology and talent.
Nvidia provides the financial resources, computing ecosystem and infrastructure needed to develop large-scale models.
Why This Matters for American AI Leadership
The geopolitical implications are significant.
The United States has maintained an important lead in advanced AI, with companies such as OpenAI, Anthropic, Google and Meta developing powerful proprietary models.
However, China has demonstrated increasing strength in open-weight AI.
That distinction matters because the most strategically important AI ecosystem may not necessarily be the company with the single best closed model.
It could be the country whose models are adopted most broadly by developers, businesses, governments and researchers around the world.
Nvidia appears to be betting heavily on that possibility.
The company’s strategy suggests that American AI leadership should include an ecosystem of accessible models rather than relying entirely on a handful of closed AI systems.
The Global AI Race Is Becoming an Ecosystem Competition
The latest Nvidia move highlights how quickly the AI industry is evolving.
The first phase of the AI boom was largely about developing increasingly powerful foundation models.
The next phase is becoming a competition over AI ecosystems.
That means controlling or influencing:
- Compute
- Chips
- Models
- Data
- Developers
- Cloud infrastructure
- Energy
- Talent
- Distribution
- Open-source and open-weight communities
Nvidia is already deeply involved in many of these areas.
Its $6 billion Poolside agreement therefore represents more than a startup transaction. It is part of a broader attempt to shape the future architecture of artificial intelligence.
Nvidia’s Strategic Bet Could Reshape the AI Market
Nvidia’s investment in Poolside demonstrates how the competitive landscape of artificial intelligence is changing.
The company is no longer satisfied with being the infrastructure provider behind the AI revolution. It increasingly wants to influence the models, software and developer communities that determine how that infrastructure is used.
The $6 billion Poolside technology and talent deal, combined with Nvidia’s separate $1 billion investment, gives the company a potentially powerful new resource for accelerating Nemotron.
The strategic objective is equally important: build a strong American open-weight AI ecosystem capable of competing with rapidly advancing Chinese models such as DeepSeek and Kimi.
For the United States, the stakes extend beyond corporate competition. AI has become a strategic technology tied to economic productivity, national security, technological sovereignty and global influence.
For Nvidia, meanwhile, the calculation is straightforward: if open-weight AI becomes one of the defining battlegrounds of the next decade, the company wants to be one of the organizations shaping that future.
The next stage of the U.S.-China AI competition may therefore not be determined solely by who builds the fastest chip or the smartest chatbot.
It may be determined by who builds the most powerful AI ecosystem—and who can persuade the rest of the world to use it.
SEO Keyword Cluster
Primary keyword: Nvidia $6 billion AI investment
Secondary keywords: Nvidia Poolside deal, Nvidia Nemotron, Nvidia open-weight AI, U.S. AI China competition, DeepSeek competitor, Kimi AI, American AI leadership, open-weight artificial intelligence, Nvidia AI strategy, Jensen Huang AI, AI model competition, U.S.-China technology war
0 Comments