Buying the Commons: Nvidia’s Reported $12.9 Billion Bid for Hugging Face

The day after Nvidia confirmed the AI boom is accelerating, a report landed that shows exactly how the winnings get spent. According to The Information, Nvidia has agreed to buy Hugging Face — the closest thing the AI world has to neutral, shared ground — for 12.9 billion dollars. If it holds, the company that already controls the picks and shovels of the AI gold rush would now own the town square where much of the mining community gathers. A quick and necessary caution first: this is a well-sourced report, not a done deal.

What Is Being Reported

Let us be precise, because precision matters here. The Information reported on Wednesday night, citing a person familiar with the matter, that Nvidia has agreed to acquire Hugging Face for 12.9 billion dollars. CNBC's source confirmed an acquisition has been part of recent talks. But no signed agreement has been reached; Business Insider and TechCrunch both report the deal could still fall apart; and neither Nvidia nor Hugging Face has publicly confirmed it. One detail is worth noting for calibration: Nvidia has moved quickly in the past to swat down reports it considers inaccurate, and its silence this time is conspicuous — suggestive, but not proof. So treat what follows as analysis of a likely and well-reported deal, not a settled fact.

The reported price is striking on its own. Hugging Face was last valued at 4.5 billion dollars in a 2023 round led by Salesforce Ventures, and it turned down a 500-million-dollar investment from Nvidia late last year that would have valued it at 7 billion, reportedly because it did not want a single dominant investor. On roughly 150 million dollars of annual revenue, 12.9 billion is a very large number — a measure of how strategically valuable this particular asset has become.

Why Hugging Face Matters

To understand the significance, you have to understand what Hugging Face is. It is, in effect, the GitHub of artificial intelligence: the central repository where developers everywhere share, download, and build on open-source AI models, along with the datasets and benchmarks used to train and test them. When Meta releases an open model, when a university publishes a breakthrough, when a startup wants to build on existing work, Hugging Face is very often where it lives. It has been the neutral, hardware-agnostic commons of the open AI ecosystem — the place that belonged, in spirit, to everyone and to no single vendor.

That neutrality is precisely what is at stake. Hugging Face matters because it was not owned by any of the giants. It was the Switzerland of AI models, and its independence was the source of its trust.

The Concentration Play

For Nvidia, the logic is close to irresistible, and that is exactly why it should concern us. Nvidia already supplies an estimated 90 percent of the world's AI accelerators. Its one long-term vulnerability is that its biggest customers — Amazon, Google, OpenAI, Anthropic — are all racing to build their own chips to escape dependence on it. Owning Hugging Face is a brilliant defensive move: because anyone downloading an open model still has to run it on hardware, and that hardware is overwhelmingly Nvidia's, controlling the world's main open-model hub helps funnel the entire open-source ecosystem back toward Nvidia silicon. It even offers Nvidia a route back into cloud computing, since Hugging Face already lets developers rent compute. Strategically, it is the act of a company using a moment of maximum strength to lock in its position for the next decade. But every reason it is smart for Nvidia is a reason it concentrates power: the dominant hardware vendor would now also own the supposedly neutral platform that steers which models the world runs.

Claim and Counter-Claim

The case that this is benign, even good, is not empty. Hugging Face's chief executive has publicly championed Nvidia's open-source efforts, and Nvidia has genuinely positioned itself as supportive of open models, not hostile to them. Nvidia has the resources to invest in the platform, keep it free for developers, and accelerate open-source AI at a time when the closed labs are pulling ahead. A well-funded Hugging Face inside Nvidia could be more capable than a cash-constrained one outside it.

The case for concern is weightier. A hardware-neutral commons owned by the company with a near-monopoly on AI hardware is a contradiction in terms — the incentive to subtly favour Nvidia's ecosystem over rivals' will always be present, whatever assurances are given at the outset. The open-source community trusted Hugging Face precisely because it was independent, and ownership by any giant, let alone the most dominant one, erodes that. There are legitimate antitrust questions about further concentrating the AI stack in one company's hands. And history with these acquisitions is not reassuring: neutrality promised at the moment of purchase has a way of quietly eroding once the integration begins. The honest synthesis: it may keep Hugging Face thriving, and it still moves a shared resource into private, conflicted hands.