The US Department of Justice’s reported inquiry into Nvidia’s proposed $12.93 billion acquisition of Hugging Face is not only a merger review involving two technology companies. It is also a test of how competition authorities understand control over the infrastructure that allows artificial intelligence to be built, distributed and operated. The questions reportedly cover whether Nvidia could use the open-source AI platform as a fundamental distribution point, while the department continues examining the chipmaker’s wider conduct across acquisitions, investments and sales practices.
That matters because Nvidia’s influence extends well beyond the sale of processors. Its chips power major AI systems, including tools developed by OpenAI and Anthropic, while the demand for those chips is driving data-centre development globally. The company has also invested in or acquired businesses connected to software, networking and computing operations. As a result, scrutiny of Nvidia reaches across the layers that make AI infrastructure function: specialised hardware, cloud capacity, software tools, research platforms and access to customers.
The reported inquiry therefore places a relatively new question before regulators. Competition in technology markets has traditionally focused on products, prices and ownership of distribution channels. In AI, the relevant distribution point may also be an open-source platform used by developers to access models, tools and datasets. The Justice Department is reportedly examining whether Nvidia’s ownership of Hugging Face could allow it to influence how competing AI companies reach users or build their systems.
Reuters reported that the DOJ’s recent questioning covers Nvidia’s acquisitions, investments, sales practices and whether the company has exerted pressure on customers. The department has also asked how the proposed Hugging Face transaction might allow Nvidia to harm rivals in the AI market. These are reported lines of inquiry, not findings of wrongdoing. The article states that government investigations are not evidence of misconduct and do not necessarily result in lawsuits.
The proposed acquisition is expected to close in the first half of 2027, according to Nvidia. The company has said Hugging Face would continue supporting work with a range of chip, cloud and AI providers. That assurance addresses one of the central concerns raised by the transaction: whether a platform that works across competing technologies would remain open after being acquired by the dominant supplier of AI chips.
Hugging Face occupies an important position in the AI development ecosystem because its platform is open-source oriented and connected to a broad range of AI work. The supplied report does not provide user numbers, market-share figures or a detailed description of the platform’s commercial structure. It does, however, identify the regulatory concern: Nvidia could gain influence over a distribution point that sits between AI developers, model providers and the infrastructure needed to run their systems.
That concern is different from a simple question about whether Nvidia is buying a direct hardware competitor. The reported DOJ questions suggest that regulators are examining the relationship between control of a leading input and control of adjacent software and developer channels. Nvidia already has the largest share of the AI chip market, according to the report. Its proposed acquisition of Hugging Face could therefore be assessed not only for what the platform is today, but also for how ownership might affect the choices available to companies building AI systems.
The issue is especially relevant to the physical expansion of AI infrastructure. Nvidia’s chips are described in the report as enabling systems such as ChatGPT and Anthropic’s Claude, while the broader AI boom is generating demand for data centres capable of supplying the necessary computing power. The report does not quantify that construction activity or provide details of individual facilities. It does establish, however, that decisions about software ownership and chip access are connected to the development of physical computing capacity.
This connection makes competition policy an infrastructure question. Data centres require land, power, cooling systems, networks and large capital commitments, but their usefulness depends on access to the processors and software that run AI workloads. If one company gains influence across several of those layers, the consequences may extend beyond technology firms. They could affect which AI businesses can obtain computing resources, which cloud and chip combinations remain viable, and how quickly new data-centre capacity can be put to use.
The DOJ’s broader inquiry shows that regulators are examining Nvidia’s position across multiple transactions. One person familiar with the matter said the department had examined Nvidia’s acquisition of SchedMD. The December 2025 purchase raised concerns among AI specialists about access to SchedMD’s open-source software, which helps run supercomputers relevant to national security. Nvidia has pledged to maintain open access and support for that software.
The department has also scrutinised Nvidia’s 2025 agreement to license technology from chip startup Groq and hire its engineers, according to the report. Officials have explored whether Nvidia offered larger allocations of AI chips when customers also purchased its networking equipment and Groq processors. That line of inquiry focuses on whether strength in a dominant product could be used to push the company into a newer market.
Nvidia’s acquisition of Run:ai, a company that makes software to help AI chips run more efficiently, has also drawn DOJ questions, the report says. Taken together, the transactions show why the regulatory examination is broader than one proposed acquisition. The department is reportedly looking at how Nvidia’s hardware position intersects with software, networking, chip optimisation and developer access.
The competitive landscape is not uncontested. Broadcom and Google are designing processors together for specific tasks and AI software, while Advanced Micro Devices has said it can produce technically superior hardware and has announced billion-dollar deals with AI laboratories and cloud computing operators. These alternatives indicate that competitors are attempting to challenge Nvidia at the chip and infrastructure level. The report does not establish whether those efforts have materially reduced Nvidia’s market position.
Nvidia’s influence has nevertheless made it a central company in the AI economy. The report describes the chipmaker as a backer of major players such as Anthropic and OpenAI and says its technology has helped spur data-centre development globally. This combination of investment, supply and infrastructure influence is what gives the DOJ inquiry significance. The central question is not simply whether Nvidia sells more processors than its competitors, but whether its commercial relationships allow it to shape the conditions under which those competitors operate.
The institutional structure of the inquiry also matters. The broad examination began under former Democratic President Joe Biden and has continued under Republican President Donald Trump, according to the report. Staff-led DOJ investigations can last for years and continue across administrations. The department also routinely seeks information from companies when evaluating proposed mergers. The reported investigation may therefore remain open without producing immediate legal action.
The DOJ spokesperson declined to comment on specific matters before the department, citing longstanding policy. The spokesperson said the department under Trump would continue prioritising the advancement of “Super Intelligence” for the benefit of the American people. The spokesperson also said it would be inappropriate for the department to conduct a merger investigation under the guise of a conduct investigation. Nvidia declined to comment, and Hugging Face representatives did not respond to a request for comment.
The political context adds sensitivity but does not establish a conclusion about the inquiry. Nvidia chief executive Jensen Huang has developed a relationship with Trump, who has publicly promoted American AI. The report says Trump has disclosed owning Nvidia stock and separately awarded Huang the National Medal of Science for advancing AI hardware. It also says there is no indication that Trump has been involved in the investigation, while critics have accused the administration of political intervention at the DOJ and the administration has disputed those accusations.
What the supplied evidence confirms is narrower and more important than any political speculation. The DOJ is asking questions about Nvidia’s proposed purchase of Hugging Face and about several other transactions and business practices. The inquiry is examining the possibility that Nvidia could extend its influence from AI chips into software, distribution and related infrastructure. Nvidia has offered assurances on continued openness in at least some of the businesses it has acquired or plans to acquire.
What remains unsettled is whether the department will seek remedies, challenge the Hugging Face transaction or take action over Nvidia’s wider conduct. The report does not provide a final finding, a lawsuit or a decision on the acquisition. The next significant milestone identified in the supplied material is the expected closing of the Hugging Face deal in the first half of 2027. Until then, the regulatory questions will continue to centre on whether AI infrastructure remains accessible across competing chips, clouds, software platforms and developers—or becomes increasingly organised around one company’s expanding ecosystem.
