The headline is about CPUs, not just GPUs
Nvidia absorbs rival CPU businesses because the company is now taking real commercial ground in a market that AMD and Intel once treated as their protected zone. The latest signal is not a clean, old-fashioned acquisition with a published price tag and a neat list of transferred assets. The current evidence points instead to a strategic absorption of business: Nvidia’s Grace CPU has passed $5 billion in trailing-12-month revenue, Vera is in full production, and management expects CPU revenue to more than double in fiscal 2028 .
That distinction matters. If this were a conventional M&A deal, the first questions would be purchase price, antitrust review, integration costs and which plants, designs or teams changed hands. Here, the sharper question is whether Nvidia is using its GPU dominance, networking stack and AI-factory architecture to pull CPU demand away from AMD and Intel without needing to buy their companies. In data centers, that can be just as consequential as an acquisition: workloads migrate, rack designs standardize, customers commit, and the dollars follow.
The October 27 snapshot shows Nvidia’s CPU story moving from future option to present business line. Grace is already a multibillion-dollar CPU franchise, Vera is positioned as the next data-center CPU in the stack, and the company’s CPU revenue guidance implies that general-purpose processors are becoming part of Nvidia’s core AI platform rather than an accessory . For AMD and Intel, that is the uncomfortable part. Their CPU advantage historically came from x86 incumbency, ecosystem depth and long procurement cycles. Nvidia is attacking from a different angle: it sells the accelerator, the CPU, the interconnect, the rack logic and the software environment as one compute platform.
Why AMD and Intel should read this as platform pressure
AMD and Intel are not being pushed out of CPUs overnight. AMD still competes in both server processors and AI accelerators, while Intel remains both a rival and a partner in some Nvidia system designs . But the strategic problem is that Nvidia does not need to win every CPU socket to reshape the economics of the market. It needs to win the sockets that sit next to the highest-value AI accelerators.
That is where Nvidia’s advantage becomes circular. Customers that standardize on Nvidia accelerators may find it easier to adopt Nvidia CPUs inside the same AI factory architecture. Once that happens, CPU decisions are no longer isolated procurement choices. They become part of a larger question: which vendor can deliver the lowest-friction path from model training to inference, networking, orchestration and power-efficient scaling?
The 24/7 Wall St. analysis frames the investor case in exactly those terms. It says Nvidia’s Vera CPU is already shipping to large customers, that CPU revenue is expected to more than double in fiscal 2028, and that the firm’s own model rates Nvidia a buy with a $260.60 price target, about 20 percent above the quoted current level . The article also contrasts Nvidia with AMD and Intel on valuation and data-center growth, arguing that Nvidia’s CPU gains strengthen its upside against both rivals .
That does not eliminate risk. The same report flags rising supply commitments, lower gross-margin guidance versus recent levels, and China restrictions in the data-center outlook as bear-case factors . In plain English: Nvidia is absorbing CPU opportunity at a time when the AI infrastructure machine is becoming bigger, more capital intensive and more exposed to geopolitics. Scale is the advantage, but scale is also the burden.
South Korea gives the CPU push a wider stage
The South Korean AI factory development is a separate corporate event, but it supports the same strategic thesis: Nvidia wants to own more of the complete compute platform. Naver said on October 27 that it had signed a computing-infrastructure agreement with Brookfield worth up to $9 billion, satisfying a precondition for Nvidia’s planned strategic equity investment in Naver . Under the structure reported by Chosunbiz, Brookfield will create a special-purpose vehicle, tentatively called Asset Co., to invest in infrastructure such as GPUs and data centers, while Naver Cloud will use the computing resources to operate the AI factory business .
This is not just another cloud announcement. The project is described as part of a 1-gigawatt-class global AI factory pursued by Naver, Nvidia and Brookfield, with Naver and Naver Cloud leading operations and Nvidia providing technology collaboration around next-generation high-performance computing platforms including DSX . Current reporting says the facility is planned to start operations at 55 megawatts in the first half of 2027, expand to 100 megawatts by the end of 2027 and reach 200 megawatts in 2028 .
That timeline is important because it shows the commitment is being converted into an execution plan in 2026, even if the first operating capacity is scheduled for 2027. The first hub is Naver’s Gak Sejong hyperscale data center, and the longer-term plan is to build toward 1 gigawatt while expanding beyond Korea into Asia-Pacific, Europe and the Middle East .
Brookfield supplies capital, Naver supplies operations, Nvidia supplies the stack
The operating model is designed to reduce Naver’s upfront capital burden while still giving it access to large-scale compute infrastructure. Aju Press reported that Brookfield will fully fund the SPV, which will secure and manage AI infrastructure, while Naver Cloud will use those resources to provide AI data-center services to customers . Chosunbiz reported that Naver plans to pay infrastructure usage fees linked to business revenue, rather than directly carrying all initial facility costs itself .
The trade-off is a major financial guarantee. EDaily reported that Naver separately decided to provide payment guarantees totaling 18.74 trillion won for Naver Cloud obligations tied to the AI factory, equal to 64.73 percent of Naver’s equity . The same report said the guaranteed amount represents the nominal total of GPU-service payments expected through the end of 2034 if the full $9 billion investment limit is used, while noting that no confirmed payment obligation existed as of the disclosure date and that actual amounts may vary .
For Nvidia, this structure creates another channel for its AI computing stack. Nvidia is expected to proceed with a roughly $1 billion equity investment in Naver after the Brookfield condition is met, and Korean reporting says Nvidia would become Naver’s third-largest shareholder once the share process is completed . That ownership angle matters because Nvidia is not merely selling chips into a project; it is embedding itself into a regional AI infrastructure platform.
The strategic consequence: CPUs become part of the AI factory
Taken together, the CPU story and the Naver-Brookfield project point in the same direction. Nvidia is turning the data center into a vertically coordinated system. The GPU remains the public star, but the CPU is becoming part of the lock-in layer. Grace and Vera give Nvidia a way to capture more value around AI servers, while DSX-style AI factories give the company a repeatable template for packaging chips, systems, software and infrastructure operations.
That is why the phrase “absorbs rival CPU businesses” is useful, even if the current public record looks more like market capture than a classic purchase. Nvidia is absorbing the growth that AMD and Intel would prefer to keep. It is doing so by making CPUs part of a broader AI platform sale, then reinforcing that platform through financing partnerships and sovereign infrastructure projects.
AMD and Intel still have strong franchises, and both remain relevant to the AI era. But the competitive map has shifted. The decisive CPU battle is no longer only about which processor is fastest or cheapest in isolation. It is about which architecture becomes the default control layer for AI factories. On October 27, the answer looked increasingly Nvidia-shaped.
Comments
Be the first to comment.