Apple is widely expected to introduce the next generation of iPhones this week, and buyers should be prepared for prices to rise. The driving force is not a brighter display or a new camera, but a dramatic surge in memory-chip costs. Soaring RAM prices have pushed component costs to levels not seen in years, and analysts say the tech industry could be living with them for quite some time.

Industry observers have coined phrases such as 'chipflation' and 'RAMageddon' to describe the moment. The memory shortage has reversed a decades-long trend in which electronics became more powerful without becoming dramatically more expensive. Company earnings calls now routinely mention memory prices and supply constraints, with hundreds of references appearing in just one quarter. The entire sector is searching for a fix, but relief is not expected soon.

The roots of the memory shortage

The common explanation is that artificial intelligence is to blame. AI systems need enormous amounts of memory, and their growth has certainly made the shortage worse. But the supply problem was already forming before the latest AI boom. It was then amplified by AI's nearly unlimited appetite for memory chips.

For years, memory manufacturers could increase supply simply by packing more chips onto each silicon wafer. Those efficiency gains began to shrink and took longer to achieve. In 2021, Micron concluded that technological improvements alone could no longer keep up with long-term demand. The company said the industry would need to process more wafers and build massive new factories.

Then the memory market collapsed. Pandemic-era purchases of computers, tablets, and phones had pulled demand forward. Once consumer spending cooled, manufacturers were stuck with excess inventory. They lost money and slowed their expansion plans. Just as the market started to recover, generative AI created a wave of demand far larger than anyone had expected.

The memory business is also extremely concentrated. Three manufacturers control roughly 90 percent of the global market. In the second quarter of 2026, Samsung held about 39 percent, SK Hynix followed with 26 percent, and Micron had 25 percent. That leaves the world dependent on a handful of companies to divide limited capacity between AI data centers and consumer devices.

Understanding the different types of memory

Memory inside a phone or computer generally comes in two forms. DRAM, or dynamic random-access memory, temporarily holds the information a device needs while running apps, loading webpages, or executing software. NAND flash memory provides longer-term storage for photos, videos, and files.

AI data centers rely heavily on a specialized form of DRAM called high-bandwidth memory, or HBM. Unlike ordinary DRAM, HBM stacks chips together and uses advanced connections to move data very quickly. It is harder to produce but much more profitable to sell. AI chipmakers and cloud-computing giants are willing to pay premium prices and sign long-term contracts to secure their supply.

HBM also consumes far more manufacturing capacity than conventional DRAM. Memory chips are made on circular silicon wafers, and a finished HBM product uses multiple larger chips stacked together. Micron estimates that producing a given amount of HBM requires roughly three times as many wafers as producing the same amount of conventional DRAM.

The financial incentives heavily favor AI. Memory makers can lock in multiyear commitments from some of the richest companies in the world instead of guessing how many phones or laptops will sell next year. Samsung has said it prioritizes customers who can guarantee committed future demand. SK Hynix has described a structural shift in demand where both AI memory and conventional memory are growing together.

AI is also pushing up demand for conventional memory. Phone and PC makers want to run smaller AI models directly on their devices, which requires more sophisticated memory and often more of it. Future models may need substantially larger memory configurations than their predecessors to support those on-device capabilities.

Why your next phone could cost hundreds more

The effect on component prices has been severe. Counterpoint estimates that DRAM prices for smartphones rose 56 percent in the first quarter of 2026 compared with the previous quarter, and then grew another 83 percent in the second quarter. For a 16GB DRAM module used in a smartphone, the estimated cost went from about $42 in the second quarter of 2025 to roughly $181 a year later, an increase of more than 300 percent.

Apple has more bargaining power than almost any other hardware maker, so its prices may not rise as much as these component costs suggest. But the scale of the increase shows how dramatically the cost of building a high-end phone has changed. Micron, SK Hynix, and Samsung have all posted enormous profits. SK Hynix recorded a 76 percent operating margin, Micron's adjusted gross margin reached 85 percent, and Samsung's semiconductor profits jumped by roughly 250 times from a year earlier.

New factories take years to build

The obvious solution is to make more memory, but that takes years. Micron is building a massive complex near Syracuse, New York. When complete, it will contain 2.4 million square feet of cleanroom space, making it the largest semiconductor manufacturing site in US history by that measure and Micron's largest facility anywhere in the world. The whole site is roughly the size of 350 football fields.

Construction involves three broad stages. Preparing the ground can take six to nine months. Building the concrete structure and facade can take another 18 months. The most complicated stage is installing the mechanical, electrical, plumbing, and piping systems needed to keep the fab running. These systems must create a space so clean that a single microscopic particle can be avoided across an enormous building.

Micron began permitting after its federal grant was confirmed in March 2024, broke ground in January 2026, and poured concrete about six months later. Yet the company does not expect meaningful output from the New York site until 2030. Its Idaho plant is further along and should begin wafer output in mid-2027. Micron is spending more than $25 billion this year, roughly double its capital expenditures from a year earlier, to speed up the process.

Samsung and SK Hynix are also investing heavily. SK Hynix is building the Yongin Semiconductor Cluster south of Seoul and moved its target completion date from 2045 to 2033. The company says it will double capacity within five years but still expects demand to outpace supply through 2030. Samsung and SK Hynix together plan to invest 800 trillion won, about $588 billion, in four new memory factories in southwestern South Korea. China's CXMT is also building additional plants, although its technology is not yet considered as advanced as the three established leaders.

How manufacturers are responding

Hardware companies cannot absorb component costs that have multiplied several times over. They can raise prices, build fewer devices, reduce memory capacity, or focus on premium products where higher costs are easier to pass on. Many are doing all of these at once.

Microsoft raised Xbox prices by another $100 to $150, leaving some models as much as $300 more expensive than at launch. It also increased the price of several Surface Pro laptops and tablets by $500 over their original starting price. Meta added $100 to the cost of its Quest 3 headset. Samsung and Google have increased the starting prices of some new phones by $100 this year.

Apple has already raised prices across its lineup of Macs and iPads. It is now expected to do the same for the iPhone. Rather than release its full lineup at once, Apple is said to be introducing the iPhone 18 Pro and Pro Max alongside a foldable iPhone this fall, while holding the standard iPhone 18 and a refreshed Air model until the spring of 2027. That concentrates the fall lineup at the most expensive end of the market, where profit margins are thickest.

According to estimates published in the Wall Street Journal, the iPhone 18 Pro could start at $1,299, which would be $200 more than the iPhone 17 Pro. Apple's then-CEO Tim Cook described memory pricing as a '100-year flood,' citing exponential increases that had already forced price increases on other products.

Apple may be better equipped to handle higher prices than most competitors. Its loyal customer base, deeply integrated ecosystem, and premium image make its phones behave more like luxury goods than interchangeable electronics. The company has even gained market share by keeping prices steady while lower-end rivals raised theirs. In markets such as China, consumers have often decided to pay extra for an iPhone when cheaper alternatives were nearly as expensive.

Other manufacturers have fewer options. Analysts report that smartphone makers are reducing expected shipment volumes while shifting toward more expensive models. If they cannot sell as many phones, they will try to get more revenue from each one. That means more OLED displays, AI features, larger memory configurations, and other upgrades that can justify higher prices. As a result, industry revenue for PCs and smartphones could stay flat or even rise while unit sales decline.

When will the shortage end?

New supply is coming, but not quickly. The memory makers are investing heavily in fabs that will take years to become operational. SK Hynix says demand will outpace supply through 2030 even after doubling its capacity. Micron says it cannot see a clear point where supply catches up because demand continues to grow almost as fast as expansion occurs. Counterpoint expects capacity to meet demand only in late 2027 or early 2028 in the best-case scenario. IDC similarly predicts the shortage will last well into early 2028.

There is another possibility: demand could collapse before supply catches up. If the AI investment bubble bursts, memory makers could suddenly face very different economics. But for now, long-term contracts between AI companies and memory suppliers make the near future more predictable. One analyst said 2027 is 'pretty locked and loaded.'

When relief arrives, it may not look like falling prices. Analysts expect the pace of increases to slow to single digits or low double digits, but slower growth is not the same as cheaper components. One industry researcher described the expected outcome as a new normal where memory prices remain 'at least triple what they used to be.'

New factories still take years to come online, AI data centers continue to consume record amounts of memory, and consumer devices need more RAM to support AI features. Memory makers also have little financial reason to prioritize cheaper consumer electronics over customers willing to pay a premium and commit years in advance. As one analyst put it, 'It's not a happy ending anytime soon.'

Source:The Verge News