Why the AI Boom Could Make Your Next Phone More Expensive

Written by Finnovate

Finnovate’s editorial team researches and creates financial content using trusted sources, regulatory references and inputs from subject experts.

Content Team

You may have no plans to pay for an AI subscription. But the next time you replace your phone, the AI boom could still affect the bill.

The connection sits inside the device: its memory chips.

Phone makers buy these components from an industry increasingly focused on supplying data centres. As spending on AI grows, those customers need more memory, and suppliers are directing scarce production capacity towards them. That puts pressure on the cost of making everyday electronics.

Samsung’s latest earnings guidance offers a glimpse of the money involved. On October 8, the company estimated operating profit of 107.4 trillion Korean won for the July–September quarter, roughly 8.8 times the figure a year earlier. These are preliminary estimates for the whole company.

To understand how that connects to a phone purchase in India, we need to follow what happens before the device reaches the shop.

The component both industries need

A phone uses memory for two different jobs.

Working memory, commonly called RAM, holds information the device needs while running apps. Storage keeps your photos, videos and downloaded files even when the phone is switched off.

These functions depend on different chip technologies, broadly DRAM for working memory and NAND flash for storage.

AI systems need memory and storage too, although their requirements differ considerably from a phone’s. They must handle large models, move data quickly and serve many requests. Micron says larger models, longer conversations and more simultaneous users are increasing the memory and storage needed to run AI workloads efficiently.

That demand reaches an industry whose factories cannot expand at the speed of software.

Why a phone maker faces competition from a data centre

Memory manufacturers must decide how to use their production capacity and where to invest next.

Samsung’s second-quarter results describe a focus on server products amid limited capacity. Higher memory prices helped its memory business achieve record earnings. The company also expected server demand to keep supply tight despite softer demand from phones and PCs.

This creates an uncomfortable situation for phone brands. Even if fewer consumers want new handsets, the components they need can remain expensive because demand elsewhere is strong.

The overlap is also becoming more direct.

A technical paper by Micron and Meta examines the use of low-power memory, historically associated with mobile devices, in data centres. Its appeal is energy efficiency. A technology useful for conserving a phone’s battery can also help reduce the power needed to run servers.

That does not mean a chip can simply be pulled from a phone and plugged into an AI server. Designs, packaging and specifications differ. But the businesses increasingly depend on overlapping memory technologies and manufacturing resources.

Why higher prices do not fix the shortage overnight

Higher prices give manufacturers a reason to invest. New factories still need to be built, equipped and brought into production.

Micron’s September earnings commentary explains that output from new memory factories takes time to increase, becoming more meaningful several quarters after production begins. The company expects supply to remain constrained in 2027 and 2028, although that remains its forecast.

Phone makers therefore have to plan products around the supply available now.

TrendForce’s September outlook projects another increase in memory contract prices during the October–December quarter: 10–15% for conventional DRAM and 15–20% for NAND flash. Its research also points to tight allocations of mobile memory as suppliers favour server applications. These are component-price forecasts, not predicted increases in phone prices.

How the cost reaches the buyer

A phone company has several ways to respond.

It can raise the selling price. It can keep the price steady and accept a smaller profit. Or it can change the product, offering a smaller upgrade in memory or other features than it otherwise would have.

The pressure can also affect which phones get made. Budget devices leave less room to absorb a higher component bill. IDC’s research describes brands cutting back on lower-priced models and shifting towards more expensive products as memory costs rise.

So the effect may not always look like a familiar handset suddenly costing more. A buyer could instead find fewer attractive options within the same budget.

There is no reliable one-to-one conversion between a chip-price increase and a phone-price increase. Memory is only part of the cost. Existing inventory, supply contracts, competition and the manufacturer’s willingness to absorb costs all affect the final price.

An older model on sale can therefore remain a good deal even while the cost of producing its replacement is rising.

India is already part of the story

An October 7 forecast from Counterpoint Research expects Indian festive-season smartphone sales volumes to fall 12% from a year earlier, while the market’s value grows 5%. It identifies rising memory costs and higher device prices as pressures on demand. These are forecasts for the season, not sales results.

Fewer phones sold alongside higher sales value is an important distinction.

A market can collect more money without selling more units. Higher prices can contribute, but so can a shift towards premium devices. That is why an increase in the average selling price should not be read as an identical price increase for every phone.

For buyers, the useful comparison remains the actual device: its price, memory, storage and other features against the available alternatives. A market-wide shortage cannot tell us exactly how much a particular model should cost.

The same boom can help one business and hurt another

Samsung illustrates both sides.

While its memory business benefited from strong demand and higher prices in the second quarter, its Mobile eXperience and Networks businesses together reported an operating loss of 0.7 trillion won. The company cited elevated component costs as a reason for weaker earnings in those businesses.

For investors, this is the more useful question behind the AI headline: where does the extra spending become profit, and where does it become an expense?

A supplier with scarce capacity may gain pricing power. A device maker buying its components may face a squeeze. Even within one company, the outcomes can differ.

That does not make a profitable chipmaker an automatic investment opportunity. The price paid for the shares, future capacity additions and the durability of demand still matter.

For the phone buyer, the connection is more immediate. The next upgrade depends partly on how memory suppliers allocate their factories and what phone brands can afford to pay. Decisions made to serve AI data centres can change the price and specifications of a device bought mainly for calls, photos and everyday apps.

Disclaimer: This article is for information and education only and does not constitute investment advice or a recommendation to buy or sell any security. Company guidance and industry forecasts may change.