Analysis
Why Nvidia's Hugging Face deal is drawing US antitrust scrutiny
US officials are asking how Nvidia's proposed Hugging Face acquisition could affect AI competition. Here's what developers should know.
The US Department of Justice has asked questions about how Nvidia's proposed acquisition of Hugging Face could affect competition in artificial intelligence, Reuters reported on October 28, citing two people familiar with the matter. The deal would bring together a leading supplier of AI chips and one of the best-known platforms for sharing models, datasets and developer tools.
The acquisition has not closed. Nvidia expects it to close in the first half of 2027, subject to closing conditions including regulatory approvals. Questions from investigators are not a finding of wrongdoing and do not necessarily lead to a lawsuit.
The deal in brief
| Question | What we know |
|---|---|
| What happened? | Nvidia agreed to acquire Hugging Face in September 2026. |
| How much? | Nvidia announced a $12.93 billion transaction. Its SEC filing breaks this down into about $11.9 billion for shareholders, subject to adjustments, and up to about $1 billion in employee retention awards. |
| What is new? | Reuters says DOJ officials have asked how the deal could affect competition and whether Nvidia could gain control of an important AI distribution point. |
| Is it complete? | No. Nvidia expects closing in the first half of 2027, subject to conditions and approvals. |
Sources: Nvidia's announcement, SEC filing and Reuters.
Why Hugging Face matters
Hugging Face is a place where developers and researchers publish and discover AI models, datasets and applications. A developer might use it to find a model, inspect its license, download its weights and compare ways to deploy it. Many projects and companies use the same infrastructure to distribute their work.
Nvidia said in September that more than 18 million developers, researchers and creators used the platform, which hosted over 3 million models and 500,000 datasets. Those are Nvidia's figures, not independently verified user or activity counts (announcement).
The distinction between an open platform, open-source software and open-weight models matters here. Being available on Hugging Face does not make every model open source or grant unrestricted commercial use. Each model and dataset has its own license and conditions.
Why competition authorities are asking questions
According to Reuters' two sources, DOJ officials have asked whether ownership of Hugging Face could give Nvidia control over a fundamental distribution point for AI and how the company might use the platform to harm rivals. The questions come alongside a broader, years-long examination of Nvidia's acquisitions, investments and sales practices, Reuters reports.
A practical way to understand the concern is to follow a developer's journey: discover a model → choose a deployment service → select hardware. Nvidia already supplies hardware used by many AI developers. Owning a major discovery and distribution platform could give it influence at another stage of that journey.
That is a potential competition concern, not proof that Nvidia has changed Hugging Face or plans to disadvantage competitors. Regulators would need evidence and legal analysis to determine whether any conduct or transaction violates the law.
What Nvidia says will stay open
Nvidia says developers will continue to choose their models, frameworks, cloud and inference providers and computing platforms. Its announcement explicitly says Nvidia hardware will not be required to build or deploy through Hugging Face.
In its SEC filing, Nvidia says it has committed to keeping the platform open in line with existing practices, allowing uploads and downloads of models and datasets and supporting other silicon vendors. Those are Nvidia's stated commitments. How they are implemented over time, and whether regulators impose additional conditions, remain to be seen.
Nvidia also argues that its infrastructure and engineering resources can help improve reliability, safety, evaluation and deployment. Those potential benefits will need to be weighed against competition concerns and the community's reliance on a platform serving many different providers.
What should developers watch?
For now, there is no confirmed change to the platform's model access, pricing or vendor support arising from the proposed deal. Developers can watch for concrete updates in four areas:
- Access: whether users can continue to find, upload and download models and datasets on the same terms.
- Hardware and hosting choice: whether competing chips, cloud providers and inference services remain supported in practice.
- Discoverability: whether search, recommendations and integrations treat comparable options fairly.
- Licenses and portability: whether individual model terms permit the intended use and whether projects can maintain copies or move their workflows when those terms allow it.
These are questions to monitor, not a list of changes announced by Nvidia. A model's license, the service's terms and the costs of moving infrastructure should all be checked before making deployment decisions.
What happens next?
The transaction remains subject to the required approvals and other closing conditions. The DOJ declined to comment on matters before the department; Nvidia declined to comment on Reuters' questions, and Hugging Face did not respond to the news agency's request for comment.
An inquiry can end without a lawsuit. It can also lead to further review or a challenge; the available reporting does not establish which outcome will occur. For developers, the question is whether a platform used across the AI ecosystem will continue to offer meaningful choice as its ownership changes.
Sources
- Reuters: US DOJ inquiries about Nvidia's proposed purchase, October 28, 2026
- Nvidia: announcement and platform commitments, September 3, 2026
- Nvidia: SEC Form 8-K, September 2026
Spotted an error in this article? Report it · Our corrections policy
