What does NVIDIA want by acquiring Groq for $20 billion?

NVIDIA caused a stir in Silicon Valley on Wednesday, agreeing to pay approximately $20 billion for a technology license from the startup Groq and hiring its core team.

This massive deal aims not only to solidify NVIDIA's dominance in the field of artificial intelligence inference computing by acquiring Groq's specialized technology but also employs a unique transaction structure to circumvent increasingly stringent antitrust scrutiny.

According to a person involved in the transaction, the deal takes the form of a non-exclusive technology license, with NVIDIA simultaneously hiring Groq's founders and executives. Although the details of the transaction have not been fully disclosed, the funding scale of about $20 billion has reached nearly three times Groq's valuation of $6.9 billion from months ago. Through this move, NVIDIA aims to gain more cost-effective and faster chip design capabilities to meet the growing demand for AI application operations.

NVIDIA CEO Jensen Huang clarified the strategic intent of this transaction in an internal email to employees. He stated that the plan is to integrate Groq's low-latency processors into NVIDIA's AI factory architecture, thereby expanding platform capabilities to serve a broader range of AI inference and real-time workloads. This means NVIDIA is attempting to fill the efficiency gap in inference chips beyond the extremely expensive high-performance training chips.

The structure of this deal mirrors the model adopted by Microsoft, Amazon, and Google over the past two years, which involves "licensing technology + hiring talent" to navigate regulatory scrutiny without formally acquiring the company.

This move not only allows NVIDIA to acquire critical intellectual property and talent but also reflects how the world's most valuable company is leveraging its cash reserves of up to $60 billion to accelerate the construction of defensive barriers in the face of challenges from competitors like Google's TPU.

Filling the Inference Gap

The core driving force behind this transaction is NVIDIA's competition for the AI inference market.

Although NVIDIA's GPUs and supporting software dominate AI model development and training, its existing chips are often too large and costly for running practical applications like chatbots (inference). The market has been seeking cheaper and more efficient alternatives, and Groq's technology was born for this purpose.

This licensing arrangement allows NVIDIA to acquire Groq's intellectual property. Groq claims that its chips outperform NVIDIA in data processing speed for specific tasks involving AI applications. In contrast, while NVIDIA's chips maintain flexibility in handling various types of operations, there is room for optimization in processing speed and latency.

Dylan Patel, chief analyst at chip consulting firm SemiAnalysis, pointed out that although Groq's first-generation chips have not yet posed a strong competition to NVIDIA, its subsequent two generations of products are about to be launched. He believes that NVIDIA may have seen a threat in Groq's next-generation technology, prompting them to take action.

The Special Structure of "Licensing + Talent Acquisition"

This transaction is not a traditional full acquisition. Groq founders Jonathan Ross, President Sunny Madra, and other employees will join NVIDIA to "advance and expand" the licensed technologyAt the same time, Groq's existing cloud business will remain in-house, with Simon Edwards, who joined in September as Chief Financial Officer, taking on the role of new CEO to continue operations.

This structure of non-exclusive licensing deals is a common tactic used by recent tech giants to circumvent regulatory scrutiny.

Microsoft, Amazon, and Google have all utilized similar frameworks to attract founders and core technologies from AI startups without formally acquiring the companies. Although Google's similar deal with Character.ai sparked scrutiny from the U.S. Department of Justice, no actual action was taken. Currently, NVIDIA is not facing antitrust scrutiny in the U.S., but it has remained cautious in describing its share in the AI chip market.

According to sources familiar with the matter, as a result of the licensing agreement, Groq's investors (including BlackRock and Tiger Global Management) will receive returns that include installment payments based on future performance. This deal is similar to NVIDIA's agreement with the networking startup Enfabrica three months ago, when NVIDIA spent over $900 million to hire the company's CEO and engineering team and paid a technology licensing fee.

The Unshakeable NVIDIA Ecosystem

Despite receiving billions of dollars in venture capital, challengers like Groq have struggled to break NVIDIA's tight control over the high-end AI chip market. NVIDIA's chips have created a high level of customer stickiness due to their proprietary CUDA programming language ecosystem.

Groq's recent business performance also reflects the difficulty of challenging the giants. The company recently lowered its revenue expectations for 2025 by about three-quarters. A Groq spokesperson stated that this was due to a lack of data center capacity in regions where chip deployment was planned, causing some revenue expectations to be pushed to next year. Groq had previously projected that its cloud business would exceed $40 million in revenue this year, with total sales surpassing $500 million.

Meanwhile, the competitive landscape is intensifying. Google's TPU is becoming a strong competitor to NVIDIA's GPU, with major companies like Apple and Anthropic using TPUs to train large models. Additionally, Meta and OpenAI are also developing their own dedicated inference chips to reduce reliance on NVIDIA. In the startup sector, consolidation trends are evident: Intel is in deep negotiations to acquire SambaNova, Meta has acquired Rivos, and AMD has absorbed the team from Untether AI.

The Cash Hoard Moat Strategy

NVIDIA is leveraging its massive cash reserves to solidify its business. As of the end of October, its cash reserves had reached $60 billion. In addition to funding dozens of cloud providers and startups that exclusively purchase or lease its chips, NVIDIA has also begun to pursue larger-scale technology acquisitions.

Previously, NVIDIA's largest acquisition was the $6.9 billion purchase of Mellanox in 2019, which has now become an important networking division for NVIDIA, contributing approximately $20 billion in revenue last quarterAlthough the $20 billion deal with Groq is not a full acquisition, the scale of the funding far exceeds previous transactions, demonstrating NVIDIA's willingness to pay a high price to eliminate potential threats and integrate cutting-edge technology in the face of increasingly specialized chip demands