Nvidia (NVDA) licensed Groq and hired its founder. The DOJ is asking if that sidestepped antitrust review.

The Nvidia logo.

Key points

  • DOJ probes whether Nvidia-Groq deal sidestepped antitrust review
  • Nvidia licensed Groq's tech and hired its founder
  • Two senators queried the same deal in March
  • Neither company disclosed a price in the joint announcement

The Justice Department is investigating whether Nvidia (NVDA) structured its licensing deal for Groq's chip technology to avoid antitrust scrutiny, the New York Times reported on Wednesday, citing two people familiar with the matter. The department opened the inquiry shortly after the deal was announced in December and has since sent Nvidia a formal request for information, according to the Times report, as carried by Reuters.

Nvidia and Groq did not describe the transaction as an acquisition. On December 24, 2025 the two companies announced a non-exclusive licensing agreement for Groq's inference technology. Founder Jonathan Ross and president Sunny Madra moved to Nvidia along with other Groq staff. Groq said it would remain independent, that GroqCloud would run without interruption, and that Simon Edwards would take over as chief executive.

Nobody has confirmed the price

The joint announcement carries no dollar figure. The reported numbers come from news organizations and they do not agree. Reuters described it as a $17 billion deal, while CNBC's David Faber reported a value of about $20 billion in December, describing it as Nvidia's largest deal on record.

The two accounts of what changed hands also differ. Groq's release says the company stays independent. Senators Elizabeth Warren and Richard Blumenthal, citing CNBC, wrote that the deal gives Nvidia "all of Groq's assets."

Why the structure is the story

Certain acquisitions above federal thresholds must be reported to antitrust regulators before closing. Nvidia and Groq instead announced a nonexclusive technology license accompanied by the hiring of Groq's founder, president and other employees. According to the Times, the Justice Department is examining whether that structure avoided scrutiny that a conventional acquisition would have received.

Warren and Blumenthal made that argument in a March 19 letter to chief executive Jensen Huang. They wrote that Nvidia agreed to pay $20 billion for the license and to hire "many of Groq's key employees, including its CEO and president," and that the deal "appears to be structured to evade scrutiny by antitrust regulators."

Their letter put Nvidia at about 90% of the market for high-end data center GPUs and 92% of the PC GPU market as of the end of the third quarter of 2025, against 7% for AMD (AMD) and 1% for Intel (INTC). It described inference chips, the kind Groq builds, as the more competitive part of the market. Other challengers there include Etched's Sohu and OpenAI's Jalapeno chip.

Nvidia used a similar license-and-hiring structure in a reported arrangement with Poolside, which has not been formally announced.

What Nvidia says

"The Groq story is a prime example of the American system working as designed to promote innovation, reward entrepreneurs, and benefit consumers," an Nvidia spokesperson said in a statement. Groq and the Justice Department did not immediately respond to Reuters' requests for comment outside regular business hours.

What the stock did

Nvidia traded at $218.05 at 10:09 a.m. Eastern on Thursday, down 2.4% from Wednesday's close of $223.67. AMD fell 2.6% to $507.41 and Broadcom (AVGO) slipped 0.5% to $362.71, so Nvidia's decline was part of a broader drop in chip stocks.

The Times reported that the agency could fine Nvidia if it finds the company mishandled the deal, though it is unlikely to seek to unwind the transaction.

Frequently asked questions

Why is the DOJ investigating Nvidia over Groq?

The New York Times reported on September 9, 2026, citing two people familiar with the matter, that the Justice Department is investigating whether Nvidia structured its licensing deal for Groq's chip technology to avoid antitrust scrutiny. According to the Times, the department opened the inquiry shortly after the deal was announced in December 2025 and has since sent Nvidia a formal request for information. Reuters carried the Times report; it did not independently confirm the inquiry.

What was the Nvidia-Groq deal?

On December 24, 2025 Nvidia and Groq announced a nonexclusive licensing agreement for Groq's inference technology. Groq founder Jonathan Ross and president Sunny Madra moved to Nvidia along with other staff, and Simon Edwards became Groq's chief executive. Groq said it would remain independent and that GroqCloud would run without interruption. The companies did not describe the transaction as an acquisition.

How much did Nvidia pay for the Groq license?

Neither company disclosed a price in the joint announcement. Reported figures differ: Reuters described it as a $17 billion deal, while CNBC's David Faber reported a value of about $20 billion in December 2025 and called it Nvidia's largest deal on record. Senators Elizabeth Warren and Richard Blumenthal used the $20 billion figure in a March 19, 2026 letter to Nvidia chief executive Jensen Huang.

Why does the deal structure matter for antitrust?

Certain acquisitions above federal thresholds must be reported to antitrust regulators before closing. Nvidia and Groq instead announced a nonexclusive technology license accompanied by the hiring of Groq's founder, president and other employees. According to the Times, the Justice Department is examining whether that structure avoided scrutiny that a conventional acquisition would have received. Warren and Blumenthal wrote that the deal appears to be structured to evade scrutiny by antitrust regulators.

More on NVDA and AMD

Dennis Singleton
Dennis Singleton

Dennis Singleton has spent years following the markets, but what keeps his attention is how AI is built. He writes about the companies behind the technology, from semiconductor designers and advanced packaging to photonics, memory, networking, and the hardware powering modern AI. His approach starts with filings, earnings, and industry research, then translates the important details into clear, straightforward analysis without unnecessary hype.