There is a quiet redistribution happening inside the global technology supply chain, and consumers are about to feel it directly in their wallets. The artificial intelligence industry's insatiable appetite for high-performance memory is crowding out supply for everyday devices, with research firm Gartner now warning that the era of the affordable PC and entry-level smartphone may be drawing to a close.
The numbers are stark. Soaring memory costs are projected to drive worldwide PC shipments to decline 10.4% and smartphone shipments to drop by 8.4% in 2026, compared to 2025 levels, according to Gartner. The underlying cause is a price surge of historic proportions: Gartner estimates a 130% surge in combined DRAM and solid-state drive prices by the end of 2026, which will increase PC prices by 17% and smartphone prices by 13%, compared to 2025 levels. Gartner's Senior Director Analyst Ranjit Atwal, speaking to The Register, described it as "the steepest contraction in device shipments witnessed in over a decade," with higher prices set to narrow the range of available devices and prompt buyers to hold on to existing hardware for longer.
The mechanism driving all of this is not a manufacturing failure or a geopolitical disruption of the kind that rattled semiconductor supply chains during the pandemic years. This time, the pressure is coming entirely from the demand side. Memory manufacturers Samsung, SK Hynix, and Micron have been prioritising production of high-bandwidth memory for AI accelerators over consumer DRAM, creating severe supply constraints for PC makers. In short, the hyperscale data centres powering the global AI boom are consuming memory at a pace that leaves very little for the rest of the market. Atwal told The Register that this distinguishes the current crunch from historical cycles, where prices rose and fell with inventory levels; here, the demand-side pressure from hyperscalers could extend the shortage through to the end of 2027.
For ordinary Australians shopping for a new laptop or phone, the most immediate consequence is the disappearance of the budget tier. According to Gartner, vendors simply cannot build a sub-$500 PC at a price that makes commercial sense. Raising the sticker price to compensate is no solution either, because buyers at that end of the market will not follow. Memory costs are forecast to account for 23% of total PC bill-of-materials expenses in 2026, up from 16% in 2025, a rise that limits vendors' ability to absorb higher component costs and reduces the viability of low-margin entry-level laptops. Gartner's modelling suggests the sub-$500 PC segment is projected to disappear by 2028.
HP's latest quarterly earnings provided a striking corporate confirmation of what the analysts are projecting. HP revealed during its Q1 2026 earnings call that memory now accounts for 35% of the cost of materials needed to build a PC, up from between 15% and 18% last quarter. HP's CFO noted that memory pricing is up roughly 100% sequentially from Q1 to Q2, with expectations for further increases later in the year. Dell and Lenovo have issued similar warnings about rising component costs and tighter margins, suggesting this is an industry-wide reckoning rather than a single company's misfortune.
The AI PC, which was supposed to democratise on-device artificial intelligence by bringing neural processing units into mainstream laptops, has become an awkward casualty of this crunch. Microsoft requires a minimum of 16 GB for its Copilot+ PC specification, and Gartner recommends at least 32 GB for new enterprise PCs. Those minimums, combined with elevated memory prices, have locked AI PCs into premium price brackets. Gartner expects that higher prices for AI-enabled PCs will delay their anticipated 50% market penetration until 2028. HP reported that AI PCs now represent 35% of its shipments, a figure that sounds impressive until you recall that industry predictions from just two years ago had these machines comprising virtually the entire market by this point.
There is a legitimate counterargument to the more pessimistic readings of this data. Some analysts point out that device replacement cycles were already lengthening before this memory shock, and that consumers holding onto hardware for longer is not necessarily a social harm. The refurbished and secondhand market is growing in response, offering a more sustainable consumption model. For enterprise IT managers, there is also a fiscal case for sweating assets rather than chasing every upgrade cycle. Gartner has indicated, though, that delayed upgrades could heighten security vulnerabilities and complicate the management of ageing devices, a concern that IT departments and government agencies in Australia will need to weigh carefully as they plan their procurement strategies.
On the smartphone side, the picture is slightly more forgiving at the premium end but genuinely difficult at the bottom. The memory-driven price surge will disproportionately affect entry-level smartphones, leading buyers to choose refurbished or second-hand models or keep their phones longer. Gartner analysts expect basic smartphone buyers to exit the market five times faster than premium buyers in 2026. For lower-income households, including many within Australia's growing cost-of-living pressured demographic, access to affordable digital devices is not a luxury concern. It has genuine implications for economic participation, education, and connectivity. The Australian Competition and Consumer Commission and consumer advocacy groups may find themselves increasingly focused on digital access equity if the budget end of the device market contracts as sharply as Gartner predicts.
For businesses and individual consumers who are weighing their options right now, the practical advice from analysts is pointed: if a replacement is needed, buying sooner rather than later is the rational move, as prices are only expected to increase through 2026 and likely into 2027. Gartner expects PC lifetime to increase by 15% for business buyers and 20% for consumers by the end of 2026 as a direct consequence of this squeeze.
The deeper structural question this episode raises is whether the AI industry's resource demands are being adequately accounted for in how governments and regulators think about digital infrastructure policy. The memory crunch is a market outcome, not a policy failure in any narrow sense. But when the upstream allocation decisions of a handful of chipmakers and hyperscale cloud providers begin shrinking consumer choice in everyday technology, it becomes a question worth examining beyond the boardroom. Reasonable people can hold different views on how much, if anything, governments should do about that. What seems harder to dispute is that the current trajectory will make the digital divide a more pressing concern, not less, over the next two years. For Australian consumers and the businesses that serve them, the time to adjust expectations is now, before prices do it for them. Consult the Australian Bureau of Statistics household expenditure data and the picture becomes clear: technology devices are no longer a trivial line item, and they are about to cost considerably more.