2026: NVIDIA Leads as the Biggest Financial Backer in AI Field, Super Unicorns Take Sides

NVIDIA has launched a remarkable capital "shopping spree" mode.

At the beginning of 2026, NVIDIA, the global giant in AI chips, initiated a remarkable capital "shopping spree" mode.

According to records from IT Juzi, from January to March, within just over two months, NVIDIA made consecutive investments in 9 leading companies in the AI field, with a cumulative investment exceeding $60 billion (including a hefty $30 billion investment in OpenAI). This series of intensive moves not only demonstrates NVIDIA's strong financial strength but also reveals its deep - seated strategic intention to transform from a "shovel seller" to an "ecosystem architect."

Investment Landscape: A "Harvester" of Super Unicorns

Information collected from channels such as IT Juzi and Jiemian News shows that NVIDIA's current round of investments features three high - profile characteristics: "high frequency, high valuation, and high concentration."

Top - tier large - model manufacturers are the top priority.

On February 27, OpenAI completed a record - breaking $110 billion financing at a valuation of $730 billion. NVIDIA and SoftBank Group each invested $30 billion, becoming one of the largest investors in this round. Subsequently, on February 12, Anthropic announced the completion of a $30 billion Series G financing, with a post - investment valuation reaching $380 billion. NVIDIA and Microsoft participated jointly. xAI, under Elon Musk, also completed a $20 billion Series E financing on January 6. NVIDIA, as a strategic investor, joined with Cisco to support the construction of the world's largest GPU cluster.

The AI infrastructure layer is the key area of layout.

On January 21, Baseten, a provider of AI inference infrastructure, completed a $300 million financing. NVIDIA alone invested $150 million, accounting for half of the financing in this round. On March 9, Nscale, an artificial intelligence data center startup, completed a $2 billion Series C financing at a valuation of $14.6 billion. NVIDIA participated in the investment, joining hardware giants such as Dell and Lenovo in betting on AI computing power infrastructure.

Comprehensive coverage of cutting - edge technology routes.

On February 18, World Labs, founded by Fei - Fei Li, completed a $1 billion financing, with a valuation reaching $5 billion. NVIDIA, AMD, and Autodesk jointly participated in the investment, targeting the cutting - edge direction of "world models." In the field of autonomous driving, Wayve in the UK completed a $1.2 billion Series D financing (valuation of $8.6 billion), and Bedrock Robotics, a construction robot company, completed a $270 million financing (valuation of $1.75 billion). NVIDIA participated in both. In addition, Runway, an AI video - generation unicorn, completed a $315 million financing in February, with its valuation soaring to $5.3 billion. NVIDIA, AMD, and Adobe Ventures jointly participated.

Track Analysis: Building an AI Empire with a Three - layer Architecture

Reviewing these 9 investments, NVIDIA's investment map presents a clear pyramid structure of "foundation layer - technology layer - application layer":

Foundation layer: Computing power infrastructure and data centers (highest proportion)

NVIDIA's most heavily invested area is AI infrastructure. Whether it's large - scale computing power consumers like OpenAI and Anthropic or infrastructure providers like Nscale and Baseten, they all directly synergize with NVIDIA's GPU sales. Nscale is committed to building vertically integrated AI data centers, covering the entire chain from GPU computing power, network equipment to orchestration software; Baseten focuses on the engineering deployment of the AI inference process. Both companies adopt NVIDIA's chip solutions. Investing in them not only locks in downstream customers but also improves the service ecosystem.

Technology layer: Multimodal and world models

AI video generation represented by Runway and world models represented by World Labs represent the evolution of AI from "understanding" to "generation" and "reasoning." Runway has established a strong brand recognition in the Hollywood and advertising industries, and its video - generation tools can directly drive the demand for high - end GPUs; World Labs explores enabling AI to have spatial understanding and physical reasoning capabilities, which requires more powerful edge computing power support.

Application layer: AI in the physical world (autonomous driving and robotics)

Wayve and Bedrock Robotics represent the implementation of AI in the physical world. The former focuses on autonomous driving, while the latter was founded by senior employees of Waymo and is committed to the automation of construction equipment such as excavators and bulldozers. NVIDIA has been vigorously promoting the Isaac robotics platform and the Omniverse simulation system in recent years. Investing in these physical AI companies helps to implement its software - hardware synergy strategy.

Investment Stage: All - in on Late - stage Super Unicorns

Different from early - stage VCs, NVIDIA's current round of investments is almost entirely concentrated in the late stage (Series D and later), and all the targets are unicorns with a valuation of over $1 billion:

● Financing rounds: Series D (Wayve), Series E (xAI), Series G (Anthropic), Series C (Nscale), all in the mature stage

● Valuation range: From $1.75 billion (Bedrock) to $730 billion (OpenAI), with an average valuation of over $80 billion

● Single - investment amount: Ranging from tens of millions of dollars to $30 billion, with significant differences in the degree of participation

This "picking the best" strategy reflects NVIDIA's unique positioning: it is not a pure financial investor pursuing financial returns but an industrial capital centered on strategic cooperation. The invested companies are all leading players in their respective tracks, with definite GPU procurement needs and scenario implementation capabilities.

Degree of Participation: From Financial Investment to Strategic Binding

NVIDIA's depth of participation in these investments can be divided into three categories:

Deep - binding type: In the investments in OpenAI and Anthropic, NVIDIA invested tens of billions of dollars. Such large - scale investments are inevitably accompanied by strategic terms such as computing power procurement agreements and technical cooperation frameworks. As stated in xAI's financing announcement, NVIDIA, as a strategic investor, will "continue to support its expansion of computing infrastructure."

Ecosystem - co - building type: In infrastructure companies such as Baseten and Nscale, NVIDIA is both an investor and a core supplier. In Baseten's $300 million financing in this round, NVIDIA alone accounted for $150 million. Such a high - proportion shareholding ensures its in - depth influence on the product roadmap.

Ecosystem - positioning type: In cutting - edge technology companies such as Runway and World Labs, NVIDIA participates as one of many investors, mainly to ensure the adoption of its own technology route and prevent competitors (such as AMD) from monopolizing the ecosystem.

Strategic Intention: From Chip Sales to Standard Setting

NVIDIA's intensive investments in this round are not simply a financial layout but are based on three strategic considerations:

First, lock in computing power demand. As AI large models enter the "trillion - parameter" competition, computing power has become a scarce resource. By investing in the largest computing power consumers such as OpenAI and Anthropic, NVIDIA has locked in GPU orders for the next 3 to 5 years in advance, forming a closed - loop of "investment - procurement - revenue."

Second, build a moat. Facing the competition from AMD and Intel in the AI chip field, NVIDIA binds the CUDA ecosystem deeply with downstream applications by investing in ecosystem companies. The technological path dependence developed by the invested companies during their growth will become NVIDIA's strongest moat.

Finally, expand revenue sources. As competition in the AI chip market intensifies, NVIDIA is transforming from "hardware sales" to "computing power operation." Investing in infrastructure companies such as Nscale and Baseten helps it explore new models such as GPU as a Service (GPUaaS) and open up new growth points in the cloud service and software service fields.

Conclusion

The investment wave at the beginning of 2026 indicates that NVIDIA is no longer satisfied with being a "shovel seller" in the AI era but is trying to become the "architect" of the entire ecosystem. From the $30 billion investment in OpenAI to the $150 million investment in Baseten, NVIDIA has woven an ecosystem network covering the entire AI industry chain with real money.

This dual - wheel - drive model of "chips + capital" may redefine the competition rules of technology giants in the AI era.

For the invested companies, obtaining NVIDIA's investment is not only a financial boost but also a ticket to enter the "NVIDIA universe." For the entire industry, it means that the AI computing power landscape is further concentrating towards the top, and ecosystem competition has become the main battlefield in the next stage.

This article is from the WeChat official account "IT Juzi" (ID: itjuzi521), author: Judy, published by 36Kr with authorization.