Ben Thompson argues Nvidia’s Groq deal puts a distinctive deal model to work

Ben Thompson argues Nvidia’s Groq deal puts a distinctive deal model to work

On January 6, 2026, Ben Thompson describes Nvidia’s arrangement with Groq as a potent example of a technology deal structured to avoid being called an acquisition. He says Nvidia is licensing Groq’s technology and hiring most of its employees. The framing matters because it points to a way a company might obtain technology and talent without an outright acquisition; the retrieved text does not explain why the parties chose this structure or what its consequences may be. Thompson calls it “the most potent application” of this deal model yet. The full analysis is unavailable in the retrieved text, so its reasoning and evidence for that judgment remain unknown.

Publication: f117494a-42a5-82da-bf50-e21e7b6ad9da

attribution — pass: The reading attributes the argument to Ben Thompson on his Stratechery page. The captured text is a single article teaser with a headline and dek, followed by subscription boilerplate. It is not a listing of several pieces, and the reading keeps each claim attributed to Thompson.

Nvidia is licensing Groq’s technology and hiring most of its employees; it’s the most potent application of tech’s don’t-call-it-an-acquisition deal model yet.

summary — uncertain: The reading flattens Thompson's evaluative stance. The headline calls the deal stinkily brilliant and promises to explain why it makes sense, so Thompson is arguing in its favour. The reading instead treats the reasoning as wholly unknown and drops that force. Smallest repair: note that the headline frames the deal as brilliant and sensible, while the full reasoning is paywalled. Groq is also never identified. The source does not explain it either, so state that gap rather than adding facts.

Nvidia and Groq, A Stinkily Brilliant Deal, Why This Deal Makes Sense

date — pass: The source dateline is Tuesday, January 6, 2026, and originalPublished is 2026-01-06T11:00 UTC, so the reading's date matches. The page ends with related-story links that may be later additions. The reading neither uses nor relies on them. The evidence is a publisher page, so no archive disclosure is needed.

Tuesday, January 6, 2026

support — pass: The factual claims about licensing, hiring most employees and the most-potent-application judgment are quoted directly from the dek. The inference about obtaining technology and talent without an outright acquisition is labeled as interpretation in the claims and stays close to the source wording. The reading states its limits and does not invent consequences or stakeholders.

it’s the most potent application of tech’s don’t-call-it-an-acquisition deal model yet.

media — pass: No media was selected, and no verified media candidates were supplied, so selecting none is correct.