Smartphone makers are under pressure due to a new global shortage of memory chips that could cost phones more in 2026. The shortage, sometimes referred to as RAMmageddon or RAMpocalypse, is not like the chip crisis of 2020 and 2023. The earlier crisis was the result of supply chain problems related to a pandemic. The current shortage is caused by the sudden emergence of artificial intelligence and the need for large amounts of memory needed to power AI data centers.
Memory chips are necessary parts of smartphones, computers, cars, and many other devices. Two common types of memory chips are DRAM and NAND. DRAM helps devices store and access information quickly. AI data centers need a more advanced type of memory, high-bandwidth memory, or HBM. HBM allows AI processors to access large amounts of data much faster, which is why HBM is critical to the design of systems such as generative AI.
And because of that, there is still limited capacity to manufacture these different types of memory. Companies such as Samsung, SK Hynix and Micron are refocusing more of their factory capacity on HBM and other high-end memory products, which are more profitable for AI companies and data center operators because they cost more to produce. As a result, the memory chips that are used in average smartphones and other consumer devices are running out of capacity.
The transition is happening as the largest tech firms race to build AI data centers. Google, Meta, Microsoft and Amazon are investing billions of dollars in AI infrastructure and are fighting to ensure they have enough chips for years ahead. They have enormous orders and long-term supply deals and hold more purchasing power than many smartphone manufacturers. Phone makers are fighting over the remaining supply, and at great cost.
Consumers are already feeling the effects. Industry estimates indicate that memory prices could rise sharply in 2026 and that smartphone shipments will fall 13 to 14 percent by 2026. The average price of a smartphone may also increase as the manufacturers pass higher costs on to buyers. Some analysts have warned that the shortage could raise average smartphone prices to record levels, and that extremely cheap smartphones may become increasingly difficult to make.
The types of smartphone companies that are predicted to face the greatest negative impact will be smaller manufacturers. For these companies, such as Transsion and Xiaomi, the only alternative may be to let their devices become more expensive, but with rising memory prices, it may become prohibitively expensive for them to maintain low- to mid-range memory. But companies such as Apple and Samsung have a better chance to focus on their low-cost devices, because, as larger companies with more negotiating leverage, they have a better chance to absorb the rising costs. There are issues even for these larger companies, though.
The memory shortage has created a block in how manufacturers are able to think about their products and new devices. The shortage is forcing manufacturers to eliminate memory and delay the release of some devices, forcing them to release only memory-intensive devices, if at all. Smartphone prices have the likelihood to increase the fastest within the greater tech ecosystem, especially since the greater tech costs are rising. It is also causing consumers to rethink their purchase strategies and switch to purchasing used devices instead of purchasing new ones.
The consequences are not only limited to smartphones. PCs, laptops, gaming devices, and cars also rely on memory chips, and these could be the target of higher prices or supply problems as manufacturers compete for fewer chips. Also, medical equipment and other industries that rely on smaller quantities of memory chips could be affected by the drop in purchasing power compared to large technology companies.
The reason may be that the construction of new semiconductor factories takes years and huge sums of money. Memory manufacturers don’t want to build too fast because they have seen several cycles of oversupply and low prices in the past. It could bring the market back down again if demand for artificial intelligence suddenly slows. For now, manufacturers focus on the most profitable products and not on increasing supply in every market.
A big part of the answer to this issue will depend on how long the AI boom continues. If companies keep investing in data centers, they will continue to need high-end memory (HBM and other advanced memory) for years to come. Many experts predict that the shortage will last until 2027, while others estimate that the shortage could be unbearable well into the early 2030s. New factories, new manufacturing processes, and more memory products can reduce the shortage over time.
The crisis illustrates the level of interdependence between the technology industries in the world today. Although AI may seem independent of the smartphone in our hands, both are powered by the same semiconductor industry. As AI companies compete to build bigger AI data centers, memory manufacturers are choosing where their small production runs will be most lucrative. For smartphone users, this means higher prices, fewer budget phones, and longer waiting times for cheaper phones. What began as an AI infrastructure boom is therefore becoming a problem for everyday consumers, proving that the race to build the future of AI can also affect the technology people already use every day.
References: