Nvidia Shifts Fears of Soaring Gadget Prices Due to AI Chip Shortage

Analysts have recently warned that the AI ​​boom could clog supply chains and drive up smartphone prices. Nvidia Using that product could make things worse.

The AI ​​data centers that tech giants around the world are spending hundreds of billions of dollars on require chips from suppliers like Nvidia, which taps a variety of components and companies to create the coveted graphics processing units.

But so do other companies AMDlike a hyperscaler google and microsoftand other component suppliers all rely on this supply chain.

Many parts of the supply chain are unable to keep up with demand, slowing production of critical components for the world’s most popular consumer electronics products. The prices of these components have increased significantly, threatening increased prices for the final product and potentially leading to shortages of some devices.

“We see the rapid growth in demand for AI in data centers creating bottlenecks in many areas,” Peter Hanbury, a partner in Bain & Company’s technology practice, told CNBC.

One of the harshest evaluations is alibaba Eddie Wu, CEO of a Chinese technology giant alibaba.

Wu’s company, which builds its own AI infrastructure and designs its own chips, said last week that it was experiencing shortages across semiconductor manufacturers, memory chips, and storage devices such as hard drives.

“There is a situation of supply shortage,” Wu said, adding, “The supply side will be a relatively big bottleneck.” He added that this could last two to three years.

Bain & Company’s Hanbury said there is a shortage of hard disk drives (HDDs) to store data. HDDs are used in data centers. These are favored by hyperscalers, i.e. large companies such as Microsoft and Google. However, as HDDs reached their capacity limits, these companies moved to another type of storage device: solid state drives (SSDs).

However, these SSDs are important components for consumer electronics.

Another big focus is a type of chip that falls under the memory umbrella called dynamic random access memory (DRAM). Nvidia’s chips use high-bandwidth memory, a type of chip that stacks multiple DRAM semiconductors.

Memory prices are skyrocketing as a result of huge demand and short supply. Counterpoint Research said it expects memory prices to rise 30% in the fourth quarter of this year and another 20% in early 2026. Even small imbalances in supply and demand can have a large impact on memory prices. Additionally, the demand for HBM and GPUs has led chipmakers to prioritize them over other types of semiconductors.

MS Hwang, research director at Counterpoint Research, told CNBC: “DRAM is certainly a bottleneck as AI investments continue to exacerbate the supply-demand imbalance and chipmakers prioritize HBM for AI.”

“An imbalance of 1-2% can cause sharp price increases, and at the moment we are seeing that number reaching the 3% level. This is very significant.”

Building capacity in various areas of the semiconductor supply chain can be capital intensive. The industry is also known to be risk-averse and did not add the capacity needed to meet the forecasts provided by the industry’s leading companies, Bain & Company’s Humble said.

“The immediate cause of the shortage is the rapidly increasing demand for data center chips,” Hanbury said.

“Essentially, suppliers didn’t produce as per the estimates provided by customers because they were concerned that the market was too optimistic and they didn’t want to overbuild very expensive production capacity. Now, suppliers have to add capacity quickly, but as we know, it takes two to three years to add a semiconductor manufacturing fab.”

NVIDIA is getting a lot of attention because of its dominance when it comes to chips that go into AI data centers.

For example, the company is a big customer for high-bandwidth memory. And its products are manufactured by TSMC, which also has other major customers such as Apple.

But analysts are focused on Nvidia’s product changes that could put significant pressure on the consumer electronics supply chain. The US giant is increasingly moving toward using a type of memory called low-power double data rate (LPDDR) in its products. This is considered more power efficient than previous double data rate (DDR memory).

The problem is that NVIDIA is increasing its use of the latest generation of LPDDR memory. This memory is compatible with Samsung and apple.

Typically, the industry only addresses demand for this product from a few large electronics companies. However, Nvidia, which has a huge scale, is now entering the market.

“We also see greater risks looming for advanced memory as NVIDIA’s recent pivot to LPDDR means it has become a customer on the scale of major smartphone manufacturers, a seismic shift in the supply chain that cannot easily absorb demand of this scale,” Counterpoint Research’s Huang said.

The link between all of these is:

From chip manufacturers such as TSMC, intel And Samsung has limited capacity. When there is great demand for a particular type of chip, these companies prioritize those, especially from large customers. That could lead to shortages of other types of semiconductors elsewhere.

Memory chips, especially DRAM, are of particular concern because they are used in so many devices from smartphones to laptops, and their prices have soared. And this could lead to price increases for popular electronic devices around the world.

According to Bain & Co.’s Hanbury, DRAM and storage make up about 10% to 25% of the bill of materials for a typical PC or smartphone. If the prices of these components increase by 20% to 30%, the total material cost will increase by 5% to 10%.

“Temporally, the impact will begin soon as component costs are already rising and the impact is likely to accelerate over the next year,” Hanbury said.

In addition to this, there is demand from companies involved in AI data centers such as Nvidia for components used in consumer devices such as LPDDRs, creating further demand in a supply-constrained market.

The world’s most popular gadgets could be in short supply if electronics companies lack the components they need for their devices or can’t get them as they move to AI data centers.

“Aside from rising costs, there’s a second problem: not having enough parts, which limits the production of electronics,” Counterpoint Research’s Huang said.

Many electronics companies are warning about the implications of all this.

xiaomiThe world’s third-largest smartphone vendor said it expected consumers to experience “significant increases in product retail prices,” Reuters reported this month.

Jeff Clark, Chief Operating Officer Dellsaid this month that the rise in component prices was “unprecedented.”

“We’ve never seen costs move at the pace we’ve seen them in the past,” Clark said on the earnings call, adding that the pressure is being felt in different types of memory chips and storage hard drives.

AI infrastructure players use chips similar to those used in consumer electronics. These are often some of the more advanced semiconductors on the market.

But there are also legacy chips manufactured by the same companies that the AI ​​market relies on. As these manufacturers focus on serving AI customers, there may be unintended consequences for other industries as well.

“Many other markets, such as automotive, industrial, and aerospace and defense, for example, rely on the same underlying semiconductor manufacturing capacity as the data center market, and these markets are likely to be similarly affected by price increases,” Hanbury said.